The future of private renting

in the UKSocial Market Foundation report

into the private rental sector

Michael Ball

SMF Private Rental Cover 22/6/04 12:56 am Page 2

The future of private renting

in the UKSocial Market Foundation report

into the private rental sector

Michael Ball

2

The Social Market Foundation

The Foundation’s main activity is to commission and publish original papers by independent academicand other experts on key topics in the economic and social fields, with a view to stimulating publicdiscussion on the performance of markets and the social framework within which they operate.

The Foundation is a registered charity and a company limited by guarantee. It is independent of anypolitical party or group and is financed by the sales of publications and by voluntary donations fromindividuals, organisations and companies.

The views expressed in publications are those of the authors and do not represent a corporate opinionof the Foundation.

ChairmanDavid Lipsey (Lord Lipsey of Tooting Bec)

Members of the BoardViscount ChandosGavyn DaviesDavid EdmondsJohn McFaddenBaroness NoakesBrian Pomeroy

DirectorPhilip Collins

Deputy DirectorAnn Rossiter

First published by The Social Market Foundation, June 2004

ISBN 1 904899 07 2

The Social Market Foundation11 Tufton StreetLondon SW1P 3QB

Copyright ©The Social Market Foundation

Designed and produced by Beaufort 5

3

Contents

CHAPTER ONEThe issues 4

CHAPTER TWOThe recent growth of market-based private renting 8

CHAPTER THREEWho rents, with what effects? 20

CHAPTER FOUROwnership and investment 28

CHAPTER FIVEQuality regulation: the re-introduction of Government control 44

CHAPTER SIXReturns and taxation 53

CHAPTER SEVENSummary of findings and policy implications 60

IntroductionThe objective of this report is to provide anassessment of the state of the private rentalsector, to present some viewpoints in relation tocontemporary debates over the tenure and tomake some policy recommendations.

The revival of market-based private rentingsince the early 1990s has engenderedconsiderable optimism. Private renting nowclearly has a positive role to play in UK housingprovision and this may have significantimplications for both economic performanceand some key social problems.

This favourable situation is at variance with thatfaced by most reports on private renting writtenin earlier decades. They had the sad feeling ofbeing a series of post-mortems on a body thatseemed dead. Any revival was widely recognisedto be a pipe-dream. Then, the impossibleproved possible, when a fortuitous combinationof circumstances arose in the space of a fewyears in the 1980s and 1990s. Policy changeoccurred at a time of favourable shifts ineconomic events and led to the creation of tensof thousands of new privately rented tenancies.All the earlier decades of pessimism have nowbeen virtually forgotten and the world ofhousing has been changed irrevocably.

The full impact of the revival of the privatelyrented sector will take many more years to workitself through the UK housing system and, as itdoes so, the consequences will undoubtedlyinfluence housing policy debate. As a result,this study can only be an interim viewpoint on along-run process of change. Even so, aconsiderable degree of uncertainty currentlysurrounds the future direction of the tenure andthe policy stances that should be taken towards it.

Rental optimists see the growth of privaterenting as heralding a major transformation ineconomic and social life by greatly increasing themobility of UK households. Governments shouldencourage its expansion as much as they can,they argue, particularly by reining in theirgenerosity to the two other main tenures, owneroccupation and social housing. Rentalpessimists, instead, fear the effects of profit-making landlords on the lives of ordinary people,unless their behaviour is severely restrained byregulation and public agencies, and worry abouta buy-to-let induced property market crash.

At the risk of spoiling the plot, it should bestated at the outset that this report will argue fora more restrained line than that of either theoptimists or the pessimists. The privately rentedsector has an important role to play in the UKhousing scene, but it can never be expected tobecome the dominant housing tenure again.Instead, it has several key and special roles andinvestors can prosper within them. Yet, the rentalmarket cannot be expected to undertake thoseroles if weighed down by excessive regulation.

Along with the growth of the new privatelyrented sector, there has been considerableexpansion of high quality research.1 Becausemuch of this work is scattered and not widelyknown in the general policy debate, the earlychapters of this report provide an overview ofthe privately rented sector and relate thefindings to the policy arenas that are theprincipal themes of interest here.

The key questionsThe argument is structured around a series ofquestions and it is worthwhile listing them atthis early stage.

CHAPTER ONE

The issues

4

• Does the UK has a small rental sector byinternational standards?

• Why has the privately rented sector grown?

• Is it likely to continue to grow or will it wiltagain?

• Would there be major economic benefits fromhaving a larger privately rented sector on thescale of some of our continental neighbours?

• Who lives in the tenure?

• Who invests in it?

• Will there be a buy-to-let crash?

• Are new controls needed on the privatelyrented sector?

• Does the tenure suffer from quality and repairproblems?

• Is the tax and subsidy regime fair?

• What would be a sensible set of reforms?

The first question is usefully dealt with straightaway: is the UK’s privately rented sectorparticularly small as a share of all housingprovision? By international standards, theanswer is that it is. This fact has formed thefoundation for many of views arguing for amuch higher role for the tenure in the UK. Yet,it is only true in the strict sense of privaterenting, not of renting as a whole.

As can be seen in Figure 1.1, few otherdeveloped countries have such a low share ofprivate renting.2 Only the Netherlands inWestern Europe has a lower share, thoughIreland’s is about the same. Interestingly, indeveloping countries, the private rental marketis often very small, including many of the highgrowth S.E. Asian economies, which rely heavilyon owner occupation.

Yet, it is not the case that the UK has a lowshare of rental housing overall, as it is in the 25-35% mid-range of countries (Figure 1.1). Thatgroup includes the world’s largest economies,the USA and Japan, and most of the moregenerally more market-oriented ones. So,ostensibly, it is not a bad group to be in.

There is a distinctive group of North WestEuropean countries with the highest shares ofrental housing, but they have distinctive, ifmixed, sets of characteristics:

• They all have rent control policies of varyingdegrees of severity and scope

• Most tenants have security of tenure

• There are large tax breaks to privatelandlords, so that they can survive in a rentcontrolled environment

• Several have large social housing sectors,notably the Netherlands and Scandinavia

• Some have large numbers of foreign nationalresidents, who cannot own and so must berenters – especially Austria and Switzerland

• Some have particular cities that are heavilyrenting oriented, because of local historicpolicies – such as Berlin, Hamburg, Vienna,Paris and Stockholm.

These characteristics of the high rental shareEuropean countries do not seem ones that is itworthwhile for the UK to emulate. Rent controlsand permanent security of tenure have beensuccessfully abolished in the UK. To go back tothem would create severe market distortionsand cost the Treasury dearly for no clear socialor economic benefit.

Where the UK is peculiar in its renting system isthe high reliance put on social rather thanprivate renting (Figure 1.2). The majority ofcountries have a two-thirds or more emphasison the private sector. Another group have a

5

roughly 50/50 split. The UK is below a third.Only the Netherlands puts a greater emphasison social housing than the UK, while Thailandseems to have no private renting at all. Thissocial housing emphasis does stronglydistinguish the UK from the other mid-rangerental share countries noted above, because allof them rely overwhelmingly on the privatelyrented sector with the exception of Finland.This means that the UK’s privately rented sectoris likely to have some distinct characteristicsfrom the same tenure in other countries.

Moreover, there will be a strong influence onparts of the privately rented sector, at least,from practices with respect to social housing.

This puts a study of the privately rented sectorin the UK in a quandary. Ideally, it would bepreferable to consider the benefits of a ‘socialmarket’ approach in the social as well as theprivate rented sector. Yet, that is bothimpractical and outside the terms of referenceof this review. So, only tangential comments willbe made about the social rented sector.

6

0

10

20

30

40

50

60

70

80

Switzerl

and

German

y

Netherl

ands

Austria

Sweden

Portug

al

Denmark

Fran

ce

Canad

aUSA

Japa

n

Finlan

d UKKor

ea

Austra

lia

New Z

ealan

d

Belgium

Greece Ita

ly

Irelan

d

Thail

and

Malays

iaSpa

in

%

All

Private

Figure 1.1 The share of rental in all housing for 23 countries3

Figure 1.2 The share of private renting in all renting for 23 countries4

However, the author does think that there is aneed for a shift in the balance between privateand social renting and for deep-seated reformsin social housing provision around ‘socialmarket’ principles that should narrow theexisting differences between the two tenures.

Outline of the reportThe rest of this report is divided into sixchapters that examine various aspects of theprivately rented sector.

The first explores the long run decline of theprivately rented sector, considers the reasonswhy it has grown again since the early 1990s andoutlines the tenures’ broad role within the UKsystem of housing provision.

Chapter 3 then looks at what sort of people rentprivate housing, concentrating on those whopay unsubsidised market rents. It also examinesarguments about the wider importance ofprivate renting in generating greater personalmobility. A further section considers the roleplayed by housing benefit.

Chapter 4 switches emphasis from the user, thetenant, to the provider, the landlord. Itdiscusses the rise of the private individuallandlord and the decline of the corporate oneand looks at management issues, the buy-to-letboom and whether buy-to-let will generate ahousing market crash. A final section deals withthe proposed introduction of PropertyInvestment Funds and their likely roles withinthe privately rented sector.

Chapter 5 examines quality and safety issues overwhich debate has been growing in recent years.In this respect, the Housing Bill currently passingthrough Parliament may represent a watershed inthe reintroduction of controls over privaterenting. There is also a discussion of repair andhousing conditions and their measurement.

Chapter 6 then examines returns and taxation. Itidentifies unique costs in private renting not

experienced in other types of housing tenure.Net returns are also described and there is adiscussion of the role of taxation, which differsbetween types of landlord as well as across thevarious housing tenures.

Chapter 7 then summaries the arguments of theprevious chapters and presents some policyproposals developed from the previous analysis.Readers wishing to obtain a quick overview ofthe arguments made in this report may wish toconsult Chapter 7 first.

1 The work of Tony Crook and Peter Kemp (Univerisitiesof Sheffield and York respectively), for example, oninvestment in the privately rented sector has beenextensive and exemplary. This report, except whereabsolutely essential, deliberately avoids the academicpractice of citing references. The Joseph RowntreeFoundation website (www.jrf.org.uk) or a good internetsearch engine provide appropriate starting places toreview the academic and professional literature.

2 The data in Figure 1.1 are only indicative of relativetenure shares, because different countries definehousing tenures in distinct ways and all facemeasurement errors in gathering statistics.

3 Countries were selected to include most developed onesand a few high growth developing ones. Sources:M.Ball, RICS European Housing Review, RICS,London, 2004; United Nations; Housing Statistics ofthe European Union; Australian Bureau of Statistics;New Zealand Statistics.

4 Own calculations from sources in footnote 3.

7

IntroductionThis chapter looks at the policy causes of thelong decline of the privately rented sector untilthe 1990s and then examines the factors thathave contributed to its revival since then. Itpoints out that the revival of private rentingoverall, in fact, has been fairly limited in termsof the total numbers of rental dwellings –especially since the mid-1990s. However, avibrant free market-based rental sector hasemerged and is increasingly taking over fromwhat was once a predominantly ‘regulatedmarket’ form of renting. With the growth of thenew free market segment, the structure andgeography of the privately rented sector haschanged dramatically, as shown below.

The tenure revolution of the twentieth centuryAt the beginning of the twentieth century, thevast majority of people rented theiraccommodation. No-one knows the precisenumber, but over 90% of households arebelieved to have rented from private landlordsjust prior to the First World War. Then the rotset in: by the early 1950s, the numbers weredown to a half; by the 1970s, to a fifth.1 By 1991,at the tenure’s nadir, only 7% of householdsrented privately. Another 2% also rented inassociation with a job or business, usually at alow or zero rent, and they confusingly are oftencombined in the data with the market-orientedprivate rented sector – giving a crude 10%approximation for private renting overall.2

Household numbers in any case are not theideal measure of the enormity of the decline inmarket-based renting over the twentiethcentury, because of problems of definition with

the term ‘household’ and changes in life styleswith much smaller average household sizes inthe modern era. Population figures, instead,bring out the decline of the private rentedsector over a three-quarters of a century periodmost starkly. Before the First World War almosteveryone rented, yet by 1991 less than 6% of thepopulation was actually living in the privatelyrented sector. So, the tenure looked desperatelyclose to disappearing altogether as an optionfor most households by the end of the twentiethcentury.

Common questions over the past few decadeshave been: what has caused the decline? Has itgone too far? What are the broaderconsequences?

At the outset, it should be clearly stated thatmuch of this decline should not be decried. Ithas been a consequence of social progress andeconomic affluence. Owner occupation hasreplaced renting as the major tenure in the UK.That tenure now contains around 70% of allhouseholds – and it has brought with it manypersonal, economic and social benefits.

In fact, people have chosen to buy rather thanto rent their prime residence to such an extentthat virtually any person that can afford to doesso nowadays. This is as true in most otheradvanced economies as it is in the UK. In part,this is because of the tax breaks and subsidieswith which governments traditionally havefavoured owner-occupation. Yet, by doing so,governments have gone with the flow.Subsidising homeownership may be condemnedby economists for generating economicinefficiencies, but it is a traditional way to winvotes because most people want to be

CHAPTER TWO

The recent growth of market-basedprivate renting

8

homeowners. They rightly believe that it is acheaper personal option (which is generally trueeven in a tax neutral world, see Chapter 6) and,in addition, homeownership provides betterpersonal security (security of tenure is generallyabsolute as long as you do not default on anyoutstanding mortgage loan) and ownershipfacilities doing what you like with your home,amongst other benefits.3

Owner-occupation in economic terms is anincome elastic good to a greater extent thanother housing tenures – so, the better-off youare the more likely you are to be a homeowner.The main criterion in household tenure choice,therefore, for most households is whether theycan prudently enter homeownership.4 Theremust, in other words, be some characteristic ofthe person or household in which they live(such as income, wealth, age or type of job) orof the rental housing supplied (particularlywhether the tenant has to pay the full marketrent) that makes people not part of a home-owning household.

This conclusion about consumer preferences isextremely important for the analysis of thefuture of the private rental sector for two reasons:

• There can be no sensible argument for goingback to the notion of a dominant free marketprivate rental sector, as in the nineteenthcentury. Countries in which the private rentalsector is currently larger than the owneroccupied one are few in number and they arecharacterised by supply-side distortionscreated by either significant rent controlsand/or large tax breaks to housing investors,as was noted in Chapter 1.

• Private rented housing in the UK has to be seenin terms of fulfilling specific roles in a system ofhousing provision in which owner occupationis the most important housing tenure.

In Britain, much of the role of a rental sector isplayed by social housing, consisting ofdwellings provided by local authorities, housing

associations and others; now generally known asregistered social landlords (RSLs). Socialhousing in the twentieth century has replacedparts of the previously privately rented stockwith dwellings that nowadays are home formainly low-income groups.

Social housing for many commentators,especially in the past, has been seen as thepreferred rental option for low income groups.With substantial political support across thepolitical spectrum, it rose to house around 30%of all households in the early 1980s. Since then,council house sales and much reduced buildingrates have lowered the tenure’s share to a fifthof households and there is now much widersupport for a larger privately rented sector. Evenso, private renting has declined markedlyamong lower income groups. Nevertheless, thecase for a long-term identifiably ‘social’ tenure,as the main rental tenure, now has less forcethan in the past.

Policy influences on the decline of private rentingBroad general social trends have been overlaidby the strong hand of policy.

1 Rent controlThe most important policy that contributed tothe decline of private renting was rent control.Rent and security of tenure controls wereoriginally introduced in 1915 as temporarywartime measures. They stayed in various guisesfor the next 75 years and the remnants of rentcontrol are only now fading away. The history ofrent control in the UK was one of a series ofcomplex but fundamentally incoherent reforms.Successive governments tried to square thecircle of maintaining supply while requiringmost rental housing to be let at below marketterms, but none dealt with the fundamentalproblems until the late 1980s.

2 Unfavourable taxesPrivate landlords for many years faced a highlydiscriminatory tax and subsidy regime meaning

9

that both owner-occupiers and social housinginstitutions received far more tax breaks andsubsidies than them.5

3 Slow strangulationFaced with long-term rent controls, an unfairfiscal treatment and a declining market for theirservices, most landlords chose the mostfinancially attractive option of selling out toeither owner occupiers or to social housinglandlords. Yet, with the strict security of tenuremeasures in place for much of the rental stock,this could only be done when a tenant quit atenancy. Given the attractiveness of rent levels,this was often only after long periods of timeand frequently only with the death of a tenant.Unscrupulous landlords would force out tenantsto gain free possession and landlords andprevious tenants could charge substantial entryfees (‘key money’) for assigning leases to newtenants in areas of high demand, such asLondon.

4 A repairs and quality disasterThe abysmal returns of private landlords alsodiscouraged them from either repairing ormodernising their properties. As a result, theremaining rental housing stock declined inquality and condition for decades.

5 Enormous public expenditure and tax drainWith no repair incentives, private rental ‘slums’multiplied. Though a proportion of existinghouses were inherently incapable of beingmodern homes, others ended up beingcondemned simply through a long-run lack ofrepair and replaced by new social housing atconsiderable taxpayers’ expense. Others weresubject to waves of ‘gentrification’ in the 1970sand 1980s, in which previously rented dwellingswere bought and improved by owner occupiers– using then highly tax efficient mortgageborrowing – and, for many years, generousimprovement grants. The resultant improvedproperties provided substantial tax-free capitalgains for their new owners; whereas if theproperties had remained in the privately rentedsector, their landlords would have been subject

to capital gains taxation if they had improvedthe properties.

The end was almost nighPrivate renting continued to collapse duringboth the 1970s and the so-called free marketThatcher era of the 1980s. Between 1971 and1989, the private rental sector shrank by 1.7million dwellings – a massive 44% over aneighteen year period (Figure 2.1). It then lookedas though the tenure might shrink to virtuallynothing.

Instead, during the 1990s, the tenure began toexpand again. The extent of the increase variesdepending on which basis of comparison isused. When the whole privately rented sector islooked at as a share of all households, thechange has been quite small (Figure 2.1), partlybecause the overall number of households hasbeen growing rapidly. When the absolutechange in the number of households and peopleliving in the more purely, ‘market’ orientedsector6 is examined, the change is far moredramatic (Table 2.1). The number of householdsrose by 27% in a decade and the number ofpeople living in the tenure has risen by 44%.The increase in people arose because therewere far less single elderly people living incontrolled tenancies in 2001 and many moreyounger sharers. On the absolute growthmeasure, particularly in terms of people, theexpansion of the privately rented sector hasbeen substantial.

Table 2.1 The privately rented sector inEngland & Wales, 1991 and 2001 Censuses

Households People1991 1.5 2.92001 1.9 4.1% Change 27 44

NB: Excludes job-related and other not available to thepublic renting

10

What has caused the recent growth ofthe rental market?Expansion of the private rental sector has beendriven by several factors. Policy changes havebeen important, but they have probably onlybeen successful because there have beensimultaneous changes in demand and supplyrelationships in the privately rented sector.

1. Direct policy reforms The most obvious change was in the policyfield. The 1988 Housing Act introducedmeasures that finally made market renting aviable option for landlords. It still enabled mostexisting controlled tenancies to remain, butintroduced new forms of renting that were moremarket-based than previous ones. The primemeasure was the introduction of the ‘assuredshorthold’ form of tenancy, under which rentalcontracts could be made at freely agreed pricesbetween landlords and tenants with security oftenure fixed only for six months. As a result ofthis new form of tenancy, landlords were finallyconfident of being able to charge market rents and of getting their property back ifproblems arose.

2 Financial liberalisationThe traditional cosy world of mortgage financeinitially faced the bracing effects of competitionin the owner-occupier loan market in the 1980s.

Yet, by the 1990s, competition had spread toother areas of housing activity. As a result,there was a gradual erosion of the previouslysubstantial interest rate premiums that mostprivate landlords had had to pay above thoseoffered to owner occupiers. With the advent of‘buy-to-let’ mortgages in the late 1990s, theprevious landlords’ interest premium andassociated credit supply constraints all butdisappeared.

3 Reforms to housing subsidies and tax reliefs Subsidies to both social housebuilding and toowner occupation have been curtailed overtime, including the gradual abolition ofmortgage interest tax relief for home owners.Housing tenures are still not ‘tax neutral’, butthe tax benefits of home ownership over privaterenting, in particular, have been lessened.

4 The rising value of housing has stimulatedrental investment

Many years of tight housing supply andburgeoning demand have meant that the UK hasexperienced one of the highest trend increasesin house prices in the advanced world since the1970s. House prices have fluctuatedconsiderably over that time, especially in realterms, in a series of boom-bust cycles. Thelatest house price boom, however, has been thelongest and strongest. It started in Southern

11

The decline and revival of private renting 1971-2002Great Britain

0

500

1000

1500

2000

2500

3000

3500

4000

1971 1981 1991 2002

Thou

sand

s

Figure 2.1 The decline and revival of private renting 1971-2002

England in the mid-1990s and still shows signsof continuing, despite repeated forecasts byexperts that it will eventually end (Figure 2.2).The effect has been to make house prices seemlike a one-way bet for almost a decade insome areas.

The prospect of investment returns hasencouraged ever more investors into theprivately rented sector. This has particularlybeen the case over the last few years, becauseof problems with alternative investmentopportunities. Stock markets, in particular, haveexhibited substantial volatility in the wake of thelate 1990s high-tech share bubble, so theprivately rented sector has looked like a betterbet. The sector, as a result, has experienced atorrent of investor funds of a character not seenfor almost a hundred years.

Many investors have had high expectations ofshort-term capital gains and, uniquely in recenthistory, have actually seen significant annualcapital gains for year after year. New landlordsconsequently have had no experience of theseesaw in property values that has traditionallycharacterised the housing market.

Such a benign market context cannot lastforever, but the years of plenty have transformed

the image of housing landlords. In contrast tothe pariahs of yesteryear, being a landlord isnow seen as a smart investment (the buy-to-letphenomenon is reviewed in more detail inChapter 4.)

5.The growth of personal sector wealth andinvestment interest

Many households have higher incomes andwealth holdings than in the past and they arelooking for somewhere to invest funds that willgive them a good return. So, there is a muchgreater pool of potential investors in the generalpopulation than would have been the casetwenty or thirty years ago. This rise in thecohort of potential landlords has resulted,amongst other things, from the higher generalstandard of living in the UK, the greaterinequality in the distribution of personal sectorincome and wealth in recent decades, the hugepool of wealth that now exists in the owneroccupied sector, the loss of traditionaloccupational pension schemes and greaterconfidence among investors that they canmanage their own long-term investment affairsrather than simply leave them to financialmarket specialists.

6.Rising housing demandHigher living standards mean that many people

12

-15

-10

-5

0

5

10

15

20

25

30

35

40

1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003

%

Real

Nominal

Figure 2.2 House price inflation 1988-2003

want to consume more housing and/or live insmaller households. This has helped stimulatethe demand for private renting as well as forother tenures. In addition, the substantialexpansion of higher education since the 1980sand the British practice of university studentsmainly living away from the parental home haveboosted a key segment of private rentaldemand.

7.Growing entry barriers to owner occupationhave raised rental demand

Another consequence of rising house prices hasbeen that they have discouraged newhouseholds from entering owner occupation.More people have no option but to remain asprivate renters than was the case in earlieryears. In 1992, 44% of house purchasers werefirst time buyers; whereas only 26% were in2002. Those that do enter owner occupation,furthermore, tend to do so at an older age. In2002, almost half as few first-time buyers wereunder 25 when compared to 1990 and morewere now in the 25-34 and 35-44 age groups.The average age of a first-time buyer is now 34,according to Halifax data.7

A disincentive to homeownership is the highproportions of their incomes that first-timebuyers have on average to devote to mortgagerepayments. The Halifax has argued that in thevast majority of English towns and citiesdwellings were unaffordable for first-timebuyers. Yet, if people can purchase, they stillfind that their outgoings are less than in rentingand they enjoy substantial capital gains thatmore than outweigh all of their other housingcosts. Their housing user cost of capital, inother words, is actually negative.

Most striking has been the increase in theaverage amount of money required to enterowner occupation, which was between 4-6 timesgreater in 2002 than a decade earlier. This cashis required to fund the non-borrowed part ofhouse prices (the ‘deposit’ or ‘equity’) and topay stamp duty, which is now charged on themajority of homes and is particularly onerous

for first-time buyers in London, the highestpriced city.

The increase in deposits has been caused byboth the rise in prices and a tightening of themortgage lending criteria.8 By 2002, the averageLondon first-time buyer was having to fund anestimated minimum deposit of £20,000,compared to only £3,000 in 1992 (Figure 2.3).This increase far outstripped the average rise innominal incomes between the two periods, sothat the share of their incomes many first-timebuyers have to devote to savings in order tofund entry to owner occupation has grownexponentially – unless parents or other relativescan help them out.

Many first-time buyers actually have even largerdeposits than this estimated minimum figure.The Halifax, for instance, suggested that its dataindicated that average first-time buyers inLondon early in 2004 were putting down almost£40,000 deposits and very high sums elsewhereas well.

Furthermore, young first-time buyers are alsomore likely to have higher levels of other typesof debt arising from previous further educationborrowings and other loans than was the case adecade ago. Such debts would additionallyinhibit the ability to save a deposit for a home.Those without very generous parental donationsto cover the costs of deposits, stamp duty andthe other expenses of buying a house andsetting up home, consequently, have littleoption but to rent privately.

Capital constraints, consequently, haveincreasingly pushed younger households intoprivate renting or forced them to remain therelonger than equivalent cohorts would have donein earlier years. Developments in the owneroccupied housing market, thus, have helped toincrease the demand for rental dwellings.

8.Capital market inefficiencies stimulate rentingThere is much talk currently about high levels ofdebt in the UK and fears about the personal and

13

macroeconomic consequences that may arise inthe future. Even so, not all debt is bad. It canhelp, for example, households to find cheaperways of living.

The arguments made above, for instance,highlight the fact that the inability to borrow tofund the costs of a deposit for home ownershiprepresents a capital market inefficiency. Thisinefficiency impacts on many householdsconstrained to rent rather than buy. This is thecase for the thousands of younger householdsthat could relatively easily cover the highercosts of borrowing implied by lower requireddeposits, especially as they are at the earlystages of their careers and can expect rapidincreases in their future incomes over a shorttime period.

It must have been particularly galling for thosethat over the course of the past five years or sohave managed slowly to save for owner-occupierdeposits in relatively poorly performinginvestment vehicles – money that could havebeen used to pay-off a cheap mortgage –simultaneously to observe the capital gains fromowner occupation rocket and the necessarydeposit stretch ever further out of reach.

The housing user cost of capital – a commontechnical measure of the true cost of housing –has for much of the past decade been far less inthe owner occupied sector than in privaterenting. Yet, younger households have been lessable to take advantage of that fact than haveother generations or their forebears by aninability to borrow. Many have been unable tosignal to lenders, the monetary authorities andpolicy makers that they could have more easilypaid back a home owner mortgage than the renton an equivalent dwelling plus simultaneouslysaving for the deposit on another owner-occupied home to which they aspired.

This point about barriers to entry to owneroccupation demonstrates that the expansion ofprivately rented housing should not beunequivocally seen as an economic and socialbenefit. There are such benefits to be sure, aswill be argued later, but the causes of thegrowth of private renting must be identified intoits component parts for the welfare gains andlosses to be properly evaluated.

A temporary rise in renter demand?To an extent, the deposit effect just examined isa cyclical one. The house price rises of the past

14

0

5,000

10,000

15,000

20,000

25,000

NE NW Mer-side Y&H E Mids W Mids East London SE SW

£

2002

1992

Figure 2.3 Deposit required by a first time buyer to purchase the average priced home(Assumes the purchaser buys an average priced dwelling at the prevailing median loan-to-value ratio.)

decade are unlikely to be repeated in the nearfuture, so that deposits might gradually becomemore affordable for potential first-time buyersand the divergence between the user costs ofthe two tenures might come more closely intoline. If the deposit barrier does weaken, it maydampen the demand for private renting.

The trend growth of house prices, however, hasbeen particularly strong over time and supplyshortages are likely to remain or intensify. In thiscase, the long-run demand for private rentingmay increase further for socially indefensiblereasons, unless financial innovations andimproved borrower screening reduce the currententry barriers to owner occupation.

The changing structure of private rentingThe revival of private renting has beenassociated with considerable changes in thestructure of the rental sector in terms of thetypes of letting, the characteristics of landlordsand the geographic location of rental property.This section explores those changes and pointsout that they are by no means complete.

One noticeable feature is that most recentgrowth in private renting was actuallyconcentrated in the early 1990s. Since then, theaggregate level has stagnated at around 2.2million dwellings in England and 2.4 milliondwelling in the UK as a whole.

When examining recent changes in private rentalhousing, it is important to recognise thatprivately renting is often treated in the availabledata as a catch-all category that is defined simplyon the basis of the housing being neither owneroccupied or socially rented. Included within it inmany official housing statistics, for instance, aredwellings given to employees as part of theirjobs, to relatives and it even includes thosesquatting. This creates problems when trying tounderstand developments in the privately rentedsector, because many statistics combined allthese disparate types of accommodation. Often,

however, a distinction is made between housingthat is available to the public (i.e. is rented to anon-employee or relative) and that which is not,which helps somewhat.

The relative constancy in the recent size of theprivately rented sector is due to two opposingforces at work. Market-based lettings haverisen, while non-market ones have fallen. Themost rapid rises in market-based lettings(mainly assured shorthold tenancies) were in theyears after their introduction in the late 1980s.Since the mid-1990s, the growth rate has settleddown to around 2% a year. By 2001/2,nevertheless, there were 1.5 million dwellings ina tenancy that had not existed slightly over adecade before. Resident landlord lettings alsoexpanded somewhat, encouraged by ‘room-to-rent’ tax reliefs, and the same is true of otherforms of no security rentals, especially holidayaccommodation.

In contrast, other parts of the privately rentedsector have declined. As rent liberalisation inthe 1980s still permitted many existingregulated tenancies to continue, there has beenonly a gradual decline in their number. Yet, by2001/2, there were only a tenth of the numberof regulated tenancies that existed during thelate 1980s and regulated tenancies nowrepresent only 5% of lettings.

Accommodation associated with employmentand other non-housing market factors has alsofallen dramatically, with almost 200,000 (40%)units disappearing in just over a decade. Job-related tenancies have declined with fallingemployment in industries, such as agriculture,mining and steel-making, where tied tenancieswere traditionally concentrated. Otheremployers have taken advantage of more readilyavailable market lettings for their staff and theyand other non-market-oriented landlords havebenefited from rising house prices and sold offtheir rental property portfolios.9

This process of switching towards a moremarket-oriented privately rented sector has

15

been associated with a lot of churn in the actualdwellings that are rented out. In 2001, forexample, over half the dwellings owned byprivate individuals had been rented out for 5years or less.

Flowing out of the privately rented sector havebeen properties sold to owner occupiers and, toa lesser extent, social housing institutions.Mainly flowing in have been existing dwellingspreviously in owner occupation. Between 1991and 2001, some 600,000 properties weretransferred from private renting to owneroccupation, with an equivalent number movingin the other direction as well.10 Some socialhousing has been converted to private renting,because council accommodation sold to sittingtenants may eventually end up in the privatelyrented sector on resale. Other new rentaldwellings have resulted from refurbishments,sub-divisions and new build.

Non-market renting is still significantDespite the fall in the non-market parts of theprivate rental sector, around a fifth of theprivately rented sector is still in non-marketforms and these parts can be expected todecline further in the future. This means thatexpansion of the market segment will have to besubstantial in the next few years simply tomaintain the privately rented sector at itscurrent relative size.

Geographical and dwelling typedifferences

Where is private renting the greatest?There are considerable geographic variations inthe importance of private renting. It is highestin England (10%). Wales has slightly less (9%),followed by Scotland (7%) and Northern Ireland(5%).11 Renting, moreover, is concentrated inparticular types of city and town, andparticularly in Southern England.

Table 2.2 lists the 25 local authority areas in

England and Wales with the highest shares ofprivate renting. Central London boroughs havethe highest shares of both market and job-related renting, reflecting both local populationcharacteristics and the extremely high houseprices there. Other London boroughs alsofeature strongly in the most private-renteroriented localities, with notable geographicvariations exhibited even within London alone.

Table 2.2 The 25 English and Welsh localauthority areas with the highest shares ofprivate renting

% of all households in the locality

Market Other Private Total PrivateRenting Renting Renting

Westminster 30 7 36Kensington and Chelsea 25 5 30Camden 23 4 28Wandsworth 22 3 25Brighton and Hove 20 3 24Haringey 20 4 24Hammersmith and Fulham 20 4 23Lambeth 18 3 21Hastings 18 3 21Oxford 18 6 24Cambridge 17 6 23Bournemouth 17 3 20Brent 17 3 20Newham 17 3 20Blackpool 16 3 19Islington 16 3 19Ealing 16 3 18Southampton 16 3 18Tower Hamlets 16 3 19Richmond upon Thames 15 4 19Barnet 15 3 18Torbay 15 3 18Manchester 15 4 19Waltham Forest 15 2 17Hackney 15 2 17Kingston upon Thames 14 3 17

Source: Census 2001.

Rounding errors lead to differences in totalsand sub-group percentages.

Only three localities in the high renter list areoutside of the South East, illustrating the

16

importance of renting in the highest house pricepart of the UK. The only other large city, apartfrom London, in the list is Manchester, whichhas an exceptionally large student population.The university towns of Oxford and Cambridgealso feature, as do a swathe of seaside towns,some of which have universities. The seasideletting market is unusual in that it is seasonal,with more holiday-makers in summer and morelower income groups in winter.

Overall, concentrations of private renting canbe summarised in the words south, big city,university and sea. In contrast, the lowest shareof private rental areas tend to be rural areas andthose old industrial areas where a tradition ofowner occupation grew up, and are concentratedin the North; where property is cheaper and,hence, owner occupation is more accessible.

What types of dwelling are rented?The typical image of a private rented dwelling isa one or two bedroomed flat. In fact, flats are ina minority. Only slightly over a third ofdwellings are flats and the remainder are mainlyterraced or semi-detached houses. Even so,flats comprise a significantly larger share of theprivately rented sector than that of the housingstock as a whole (16%). Most are relativelyspacious. Few are overcrowded on the bedroomstandard and three-quarters of lettings areunfurnished.12 Only a fifth have no access to agarden or terrace.

The stock is significantly older than in othertenures. 40% was built before 1919 and almost60% before 1945. This reflects the fact that mostprivately rented properties are located in themore central parts of the UK’s towns and cities.

A segmented privately rented sectorThe data on tenancies and the geography of theprivately rented sector highlight the key divideswithin it that make talk of a single sector rathermeaningless. The privately rented sector whentreated as a whole is as much an artefact ofstatistical convenience as an expressionof reality.

Regulated tenancies are different from themarket sector. There are separate marketsegments for holiday lettings, lodgings,furnished and unfurnished accommodation toname but a few. Some localities have quite adense privately rented market. Others have onlya handful of rented properties, which makes ithard to talk of a market in a meaningful sensebecause so few properties are traded as rentaldwellings within a given time period. Thisabsence of a rental market is more likely to becaused by demand, as households in thoseareas prefer other tenures, than supply, whereno-one is prepared to be a landlord at currentfree market rents.

Segmentation exists in even the moreconcentrated rental localities because ofdifferent types of client – the rich and studentsin inner London to name but two. Somesegments in such localities will also representthin markets.

Further divisions in the privately rented sectorare added by tenants on housing benefit.Benefit rules generate a distinct market segmentbecause they influence the setting of rents.Emergency ‘bed-and-breakfast’ accommodationfor those classified as homeless is anotherquasi-social part of the privately rented sector.

Private renting is a residual type of housingcategory as well, adding further segments. Suchresidual features stretch across widely differenttypes of housing. Dwellings that form part of alegacy, for example, may be rented out whenbequeathments need to be sorted out. Similarly,nominally ‘rented’ property – though often inpractice vacant – frequently is very dilapidatedhousing or that located in areas of very lowdemand now common in rundown areas ofnorthern English towns.

Renting in each of those situations may betemporary while owners sort out their options,wait for compulsory purchase, or come to termswith holding a worthless asset. Or renting maybe simply a statistical artefact, because all

17

housing must be classified under one form ofhousing or another.

Such odd ball cases help to explain therelatively high vacancy rate in the privatelyrented sector, which at 6% is twice as high as inother tenures. Around half, furthermore, arewhat are termed ‘problem’, or low quality,vacancies. Turnover vacancies, however, are alsohigh. They are three times as high as in owneroccupation and 50% higher than in socialhousing, for reasons that are explained inChapter 6.

When discussing the privately rented sector, it isimportant to recognise this myriad of segments,because what characterises one segment cannotbe said to apply to others. The dataunfortunately often ignore such nuances, so thataverages across these divides are all that areavailable. It, nevertheless, is important to try anddistinguish between these sectors through thestatistical fog, because it tends to colour andcloud debate over the privately rented sector.

ConclusionThe 1990s witnessed a sea change in theprivately rented sector with a dramatic growth inmarket-based renting. This growth waspredominantly on the basis of the assuredshorthold tenancy, which is now the defaulttenancy arrangement unless leases stateotherwise. Expansion was facilitated bylegislative reform in the late 1980s, financialliberalisation, a revival of the small landlordinvestor, and increased demand amongstyounger cohorts of the population thatincreasingly found it difficult to enter owneroccupation.

Private renting is geographically concentrated inthe South of England, although it operates atvarious scales throughout the country. Thetenure is particularly concentrated in largecities, university towns and seaside resorts.Most lettings are of houses, but with a greaterpreponderance of flats than in the housing stock

as a whole. The stock of rental housing isnoticeably older than average, reflecting theinner city nature of much of this tenure. Rentalmarkets are segmented, so generalisations haveto be made with care.

Most of the growth in the privately rentedsector and assured shorthold tenanciesoccurred in the first half of the 1990s and thetenure has expanded little since then, so thatthe tenure seems to have stabilised at around10-12% of the overall housing stock. There hasbeen some growth in market-based renting andfurther declines in non-market-based rentingsince the mid-1990s, though change has beenslow. The number of households in market-based renting seems to be rising on trend byabout 2% a year. With non-market-basedrenting still a fifth of the rental stock, theprivately rented sector could still decline interms of its overall tenure share in the nearfuture.

Nevertheless, the revival of the private rentalover the past 15 years has been one of the mostspectacular examples of market forces helpingto generate an important, successful socialchange in Britain within a relatively short periodof time. One downside to this conclusion is thatsome of the extra demand for private rentinghas come from younger households that havebeen excluded from the benefits of owneroccupation by the rising cost of entering thattenure.

1 Source: ODPM.

2 1991 Population Census data. Devolution has had theunfortunate side effect of balkanising UK housingstatistics, so that this section and other chapters mainlyrefer to data for England only. Most of the private rentalsector is in England, however. All data cited can befound on the ODPM website, unless otherwise stated.

3 Less realistically, many people periodically believehousing is the best investment in the world – thoughthis affliction is as great for potential landlords as it isfor owner-occupiers.

4 This viewpoint, for example, is now coming topredominate in discussions of tenure choice in the USA,see Jones, L.D. ‘Testing the central prediction of

18

housing tenure transition models’, Journal of UrbanEconomics, 1995, and Gyourko, J. ‘Access to homeownership in the United States: the impact of changingperspectives on constraints to tenure choice’ inO’Sullivan, T. and Gibb, K. Housing Economics andPublic Policy, Oxford, 2003.

5 Housing economists for years pointed out the futility ofsuch policies, but were generally ignored.

6 This is defined here as the Census data on the rentingprivately furnished and unfurnished sectors in 1991 and therenting from a private landlord or letting agency in 2001.

7 Halifax data tend to differ from ODPM data, becausethe Halifax only has a large, but still limited (around afifth), share of total housing market transactions.

8 Average loan-to-value ratios declined from 95% to 89%for first time buyers between 1992 and 2002 accordingto CML data.

9 The decline of some other industries, like local pubs,would have also contributed to the loss of employer-related accommodation.

10 English Housing Condition Survey.

11 The population census records a higher share for rentalhousing than these government data at 12% for Englandand Wales in 2001.

12 Data from Survey of English Housing provisional2002/3.

19

IntroductionThis chapter looks at the types of people whorent houses, concentrating on the largerunsubsidised sector but also considering theimpact of rent subsidies in the form of housingbenefit. It examines as well the arguments putforward that the privately rented sector provideswider economic and social benefits in terms ofsuperior opportunities for household mobilitythan other tenures.

A highly mobile group of peoplePrivate renters, on average, are a highly mobilegroup (Table 3.1). A third of them move within ayear – with getting on for a half of tenants1 inthe furnished sector staying for less than a year.Another 30% or more move within three years.This means that altogether over 60% of tenantsmove within 3 years and, as these data includethe regulated and non-market sectors, it isreasonable to assume that the vast majority ofmarket renters move within 3 years in both thebenefit and non-rent subsidised sectors.

Mobility is phenomenal in comparison to othertenures. For instance, only 6% of those owneroccupiers with mortgages and 12% of socialtenants move within a year. At the otherextreme, over half of home owners and 40% ofsocial tenants stay in the same accommodationfor more than ten years.

Where do new tenants come from?Recent movers in the privately rented sector areeither creating a newly-formed household ormoving from one of the three tenures: owneroccupation, social housing and private rentingitself.2

A fifth are newly formed households. Around15% come from owner occupation, many ofwhom previously had mortgages. They maymove into renting as a stopgap in order to givemore time to search for a suitable owneroccupied dwelling. This type of move hasprobably increased with the rise in stamp duty,which now imposes a large penalty on homeowners that move frequently within owneroccupation. Alternatively, owner-occupiers maymove into renting due to a family break-up orbecause they shift from a cheap to an expensiveregion. A much smaller group come from socialhousing (6% of recent movers in the privatelyrented sector). Some social tenants move intothe privately rented sector because they can getbetter accommodation by doing so. Others areejected from the social sector as bad tenants.When letting to such latter ex-social tenants,private landlords may well be unaware of theirpast behaviour as tenant behaviour informationis not freely available.

All such entry moves into the privately rentedsector, nevertheless, constitute less than half of

CHAPTER THREE

Who rents, with what effects?

20

Table 3.1 Length of residence by housing tenure

Length of residence in yearsless than 1 1-3 3-5 5-10 10 & over All

% shares in 2002 unfurnished 30 31 11 10 17 100furnished 46 32 9 6 7 100

All rented privately 34 31 11 9 15 100All tenures 10 17 11 16 46 100

Source: Labour Force Survey

all moves into the tenure and they are matchedby a virtually equal number of moves in theopposite direction. Over half of privately rentedsector moves, therefore, are simply existingtenants moving from one private tenancy toanother.

A majority of within-the-same-tenure moves isalso the common experience in the other twomain tenures: owner occupation and socialhousing. Most of the extra churn, or mobility,within the privately rented sector, therefore,does not arise through a markedly higherproportion of moves from other tenures into itcompared with other tenures, but because ofexisting tenants moving frequently within theprivately rented sector – often in the samelocal area.

Mainly young peopleMost people who rent are young. This isparticularly the case when the tenancy has areference person paying the full-market rentwithout assistance from housing benefit. Over aquarter of those tenancies have a referenceperson aged under 25, many of whom arestudents.3 Another 37% are aged between 25and 34. So, taking these two groups together,almost two-thirds of non-rent subsidisedtenants are younger people. That is a far higherproportion than the share of these age cohortsin the population as a whole.

If only assured shorthold tenancies areexamined – the real market sector – theyouthfulness of tenant households is probablyeven higher. Unfortunately, there is not anadequate breakdown in the available data todemonstrate this, but it is reasonable to assumethat many mixed households with no directfamily ties consist solely of young people.Making inferences from that information wouldsuggest that the overall share of people age 34and under living in the non-rent subsidisedmarket part of the privately rented sector couldeasily be three-quarters or more of the totalpeople living within that category. This youthful

characteristic of the privately rented sectormarks it out noticeably from the other housingtenures.

Other countries’ privately rented sectorssimilarly have larger shares of young people inthem, no matter the size of the share of privaterenting overall. Such a younger than average ageprofile of tenants, for instance, is common inthe United States and Australia, particularly inthe more expensive parts of the market, which isthe best comparison because social renting inthe UK houses many more low-income tenantsthan in it does in those two countries.

European countries with large private rentalsectors, such as Germany and France, do have abroader range of age cohorts living in privatelyrented accommodation, though even thereprivate renting still has a young household bias.The greater age range in those countries,furthermore, arises for special reasons in that itis associated with rent controls and security oftenure provisions of varying degrees ofstrictness. These encourage people to stay inthe tenure longer, but militate against mobility,especially in areas of high housing demand.

When looking at housing tenures in terms oftheir importance for each adult age group withinthe population as a whole, the private rentalsector is the most important housing tenure inthe UK for under 25 year olds (Figure 3.1). Forolder age cohorts, owner occupation takes overthe lead role, although a fifth of 25 to 34 yearolds still live in the privately rented sector;whereas, amongst the 35 plus age groups,private renting is progressively under-represented with relatively small numbers in thetenure, especially in modern assured tenancies.

Younger households also pay the highest rentson average (Figure 3.2). Average non-subsidisedrent levels, in fact, decline consistently as theage range increases. This is because youngerhouseholds are more likely to live in the mostexpensive places, such as London and universitytowns. They often absorb the greater cost of the

21

rents they pay by sharing with others.

Household typesThe majority of reference person tenants are inwork (54% full-time and another 8% part-timein 2002 according to the Labour Force Survey).This contrasts with social renting, where lessthan a quarter of reference persons are in full-time employment and over two-thirds areeconomically inactive. This still leaves asignificant third of privately rented sectorhouseholds with an economically inactivereference person. Many of the latter will be inthe non-market sectors and others have their

rents subsidised by housing benefit.

The household types that are most stronglyrepresented in private renting are those sharing,lone parent and single person households;whereas couples with or without children areunder-represented. To an extent this reflects,both the age distribution of tenants and theentry rules for access to social housing.Because of the uneven distribution of the totalpopulation between household types, however,their relative importance as a share of privatesector tenants is somewhat distinct, as Figure3.3 shows.

22

0

10

20

30

40

50

60

70

80

90

<25 25-34 35-44 45-54 55-64 65-74 75+Age

% of age cohort living in each tenureOwner occupiers

Private tenants

Social tenants

%

Figure 3.1 % of age group living in each of the three principle tenures*

*Tenure breakdown for each age cohort sums to 100%.

Figure 3.2 Median weekly rents by age of tenant 2001/2

Couples, with or without children, are thelargest group at almost two-fifths ofhouseholds, followed by single-personhouseholds at a third, sharers at around a fifthand lone parents at around a tenth.

The distribution of income in the privately rentedsector most closely represents that of the overalldistribution of income as a whole, compared tothe other tenures where owners tend to be betteroff and social renters worse off (Figure 3.4). Evenso, there are somewhat more lower incomeprivate tenants than in the population as a whole,but that may reflect the relatively early stages atwhich such tenants are in their life cycles and

ignores the future earning power of many.

Will there be enough youngerhouseholds around to sustain private renting?As younger households are the mainstay of theprivate rental market, it is worth examiningrecent population forecasts to see how robustdemand is likely to be in the future. Populationforecasts are an inexact science and do not fullyincorporate economic influences which areparticularly important with regard to the tenurelocation of future housing demand, but they doprovide some interesting signals.

23

0

5

10

15

20

25

30

35

40

Couple no kids Couple kids Lone parent Sharing One person

%

0

5

10

15

20

25

30

Under 100 1-200 2-300 3-400 4-500 5-600 6-700 700 & over

private rentersall households

£ per week disposable income of reference person and partner

% o

f all

cate

gory

Figure 3.3 Types of households: % share of each type out of all households living in private renting

Figure 3.4 Income distribution of private renters 2001/2

The message from population forecasts is thatin many areas of currently high rental demandthere is going to be a significant fall in theabsolute number of people in the core 15 to 44year old age groups, the current mainstay of theprivately rented sector. These age groups areonly expected to fall as a proportion of theEnglish population, by about 3% between 2002and 2021, which does not imply any great threatto the overall pattern of rental demand. Yet, inLondon and other parts of the South-East, theimpact is expected to be far greater, with a lossof half a million people of that age groupbetween 2002 and 2016, 11% of the 2002 figure– because the middle range of the age pyramidwill be squeezed from both sides by a growingnumber of the young and the old. As privaterenting is so important amongst this age groupin the metropolis, this shift in age structurecould lead to a significant downward pressureon private renting demand, unless it is offset bysignificant increases in other sources of demand.

Chapter 2 has already highlighted the sharpdecline in the non-market element of theprivately rented sector and pointed out that themarket sector will have to grow significantly tosustain the privately rented share of housingprovision. The demographic forecasts areanother indicator that there can be littleexpectation on current trends and forecasts of a major growth in the privately rented sectoras a whole.

Mobility and the privately rentedsector The high mobility of the privately rentingtenants is often argued to be the tenure’snumber one economically and sociallybeneficial characteristic. There are several typesof benefit:

1 Labour mobility It is often claimed that labour mobility isimproved, as the rapid churn of rental tenanciesmeans that there is more likely to beaccommodation available for someone wanting

to move into an area in the privately rentedsector than elsewhere. Transactions costs forrenters are low, compared to owner occupation.Time costs may also be greater in the latter,given the time required to buy (and possibly sellin a chain) an owner occupied dwelling. Timecosts are also high in social renting associatedwith form filling, interviews and queuing.

It has also been suggested that higher levels ofprivate renting lower long-term unemploymentrates because of this mobility effect. Researchpublished in the mid-1990s, for example, arguedthat a 5% increase in a countries’ home ownershiprates raised unemployment rates by 1%.4 Laterinvestigations using more sophisticated modelsand micro-data, however, found a much weaker ornon-existent relationship.5

2 Tenure of transition Mobility makes the tenure play a useful role as atenure of transition. This may be a life cycleeffect, say, with younger households movinginto the tenure and leaving it when they areolder; or due to household breakdown, causedby divorce, separation or illness; or because ofhomelessness and poverty when no socialhousing is currently available.

3 Shock absorberMobility can also be seen in a more aggregateway as part of a process of system adjustment. Itcould be argued that private renting enables thehousing system of a local economy to adjustmore easily to disequilibrium shocks – throughits labour force and transitional demand rolesand because of the potential inward andoutward flows of investment capital withvariations in contemporary returns. Lesstechnically, it is like the low gear of a car,helping the housing system get over difficultbits by providing the flexibility and momentumto facilitate change.

The problems of first-time buyers in enteringhome ownership during the current price boom,discussed in the previous chapter, is a case inpoint. When house prices moderate and come

24

more into line with first-time buyer savingabilities, younger people might end up moreoften buying than renting.6 The flexibility of therental system enables such adjustments to occurfar more easily than otherwise. The first-timebuyers of the late 1980s boom, in comparison,did not have such a flexible rental sector to fallback upon as prices temporarily rose and,hence, many more of them entered owneroccupation with unrealistic amounts of debt andsubsequently defaulted on their loans.

Care must be taken with such mobility claims,however, as they may forget aggregationproblems, because what works at the individual,or micro, scale cannot be assumed to work atthe aggregate, or macro, scale. Many currentarguments about mobility and the privatelyrented sector fall foul of this issue.

Two examples illustrate aggregation problems inthe context of the current debate over mobilityand private renting.

1 Relative tenure size. It may be true thatturnover is much higher in private renting thanowner occupation, but it is the relative scale ofthese two tenures that is more important whenexamining the ability of anyone to move into alocality. As owner occupation is a much biggertenure than private renting, there will usually bevastly more properties available to buy in alocality than to rent. So, owner occupationoffers more choice as well as opportunities tomove into most parts of the country than doesprivate renting. Only in areas of the country andsegments of the housing market where privaterenting is plentiful can the rental sector providelarge numbers of potential transactions, such asin the Inner London flat market.

Phrases like ‘we need a larger private rentalsector to encourage mobility of the workforce’fail to recognise this relative size issue. The vastmajority of the workforce lives in owneroccupation – so, it is more likely to be true tosay that the country needs more owner occupieddwellings to encourage labour mobility than

rented dwellings. A larger proportion of owners,furthermore, move for job-related reasons inEngland than do private tenants.7

Similarly, to say that ‘more privately rentedhousing is needed to deal with theconsequences of the increasingly commonoccurrence of family break-up’ are potentiallywide of the mark. Most housing movesassociated with divorce or separation arise inowner occupation. A recent study of newhousehold formation caused by divorce, forexample, estimated that about a fifth of the newhouseholds created by divorce move into privatetenancies supported by housing benefit, butthat far more move into social housing andowner occupation. In fact, given the scale ofowner occupation relative to the two rentaltenures and the age profile of owner occupiers,separation and divorce was mainly a withinowner occupation event.8 So, even though theprivately rented sector plays a significant role,more affordable owner-occupier dwellingsmight be of greater numerical importance indealing with the housing issues surroundingdivorce than private renting.

2 Supply-side constraints The highermobility characteristics of private renting,whatever the scale of the tenure, cannotincrease the size of the housing stock in alocality and that is what determines how manypeople can live there.9 More rental housing byitself will not, for example, make it easier fornurses and policemen to find somewhere to liveat lower cost in areas of high demand. Onlyextra housing supply can do that, which takes along time to build and to have a market impact,even when it occurs. Similarly, few extra peoplecould move into a booming labour marketunless they induced others to vacate theiraccommodation and move out of the same areaand they could equally be active in the labourmarket as well. More private renting by itself,consequently, cannot deal with housing-inducedlabour shortages. The aggregation problem hereis encapsulated in the cliché ‘like movingdeckchairs on the Titanic’.

25

There is also a composition problem as well asan aggregation one in much current debateabout the privately rented sector. A distinctgroup of mainly younger people and a sub-sector of low-income groups on housing benefitor in rent-regulated accommodation live in theprivately rented sector and it can be dangerousto generalise from their housing experiences tomore general economic and social problemswith key links to housing.

Private tenants and housing benefitHousing benefit is paid directly to landlords tooffset the rental payments made by tenants. Inthe majority of cases, it pays all or most of atenant’s rent and cost £9.5 billion in Englandalone in 2000/1 for all types of tenure.

Around a fifth of private tenants are on housingbenefit (22% in 2002/3). Though this is asignificant number, it is far less than the 60% ofprivate tenants on benefit in 1987.10 In contrast,the share of social tenants receiving housingbenefit has stayed pretty much the same ataround two-thirds of all tenants. Private tenantson housing benefit pay significantly lower rentsout of their incomes than do those not receivingbenefit (see Figure 3.2 above). Yet, even so,they still pay a quarter of their median incomein rent on average, compared to only 6% in thesocial sector.

The decline in the role of housing benefit in the privately rented sector reflects thesubstantial transformation of the tenure over the past 15 years that has been mapped outabove and elsewhere, plus the overall decline in households on housing benefit, asunemployment has fallen and benefit rules have been tightened up.

Tenants on benefit are unpopular with manylandlords, with only around 20% being happy toserve that market, and many advertisements oftenancies explicitly state ‘no benefit’, whichmust seem strange to the uninitiated. Thereasons for the unpopularity are threefold and

relate to landlord costs, risks and theconsequences for their returns:

1 Benefit tenants are regarded as more of a riskin terms of anti-social behaviour, damagingthe property and generating additional repairsand management effort.

2 Perhaps more important is that, althoughhousing benefit is generally paid directly tolandlords, it is paid in arrears once theadministrative system has decided how muchhousing benefit a tenant is eligible for, andwhether the rent charge is reasonable, whichcan take weeks or months to process.Moreover, if a tenant’s circumstances change,the whole process has to be gone throughagain. This can lead to landlords having torepay benefit payments, when they arereduced, and then attempt to recover thedeficit in rental income from the tenantdirectly.

3 Finally, there is a degree of rent controlarising from the administration of housingbenefit, which many landlords dislike. Toavoid the threat of collusion betweenlandlords and tenants over the setting ofrents, in the early 1990s a local ‘referencerent’ system was set up, whereby rents shouldnot be out of line with typical local rents.Reference to the rent officer service todetermine a rent can be made by the housingbenefit authority, usually a local authority or aprivate agency to whom the task has beenoutsourced, when the rent is thought to beout of line. Roughly, 30% of private tenants’rents associated with housing benefit arereduced in this way.

This laudable attempt to avoid fraud throughlandlord-tenant collusion unfortunately ends upinfluencing local market rent levels. Whendemand is increasing in a locality, rents can beexpected to rise, but this will make them out ofline with current reference rents. Rent officersare trained in the art of taking the ‘scarcityelement’ out of rents. This process is a mystery

26

to economic theory, where all goods andservices in order to command a non-zero pricehave to be scarce, but has crept back intoprivate renting in the UK through themachinations of the housing benefit system.This process has reintroduced a de factoelement of rent control back into the privatelyrented sector.

In so far as landlords’ rents are kept below freemarket levels they will respond in order to keepup returns to a competitive level in the way inwhich they did under earlier rent control: eitherby quitting the market, which helps to explainthe decline of the benefit-related privatelyrented sector, or by cutting quality. Housinglobbyists and many left-leaning politicians decrythe evil, exploiting landlord as a result ofquality problems for housing benefit tenantsand campaign for more regulation and qualitycontrol. By doing so they ignore the economicsof the situation they are trying to confront,which suggests that the benefit rules havecreated perverse incentives for landlords thatlead to quality problems.

The government is trying to deal with some ofthese issues in a current small-scale experimentthat offers new housing benefit recipients inPathfinder local authority areas a standard LocalHousing Allowance voucher, which they can useto shop around for accommodation. This givestenants an incentive to search and negotiatewith landlords and initial results indicate thatthis is occurring. Yet, the introduction of astandard local allowance does not alter the factthat a certain sum of money will pay the rent ofaccommodation for a particular size and qualityat a specific point in time. How is a standardallowance to be adjusted to take account ofchanges in rents arising with variations in localhousing demand and supply balances and priceinflation? Again, the presumption is that thehousing benefit authorities will know what aproper ‘fair’ rent should be. The standardallowance will thus have similar consequencesto reference rents in setting de facto local rentlevels for private tenants on benefit.

1 Strictly these data refer to ‘reference persons’ – theperson that the survey interviewers talked to in thehousehold. In earlier days, they would generally havebeen identified as the head of household

2 The data on recent movers is based on Census 2001information.

3 Campus accommodation is excluded from the definitionof the private rented sector. A recent study hassuggested, however, that half the full-time UK studentpopulation rents privately. As tenants, they are popularwith landlords as they are generally well-behaved andreliable payers (Joseph Rowntree Foundation FindingsD60, 2000).

4 Oswald, A.J., The housing market and Europe’sunemployment: A non-technical paper, WarwickUniversity, 1999.

5 See, for example, Green, R.K. and Hendershott, P.H.,Home ownership and the duration ofunemployment: A test of the Oswald Hypothesis,mimeo, 2001 and P. Flatau et al. Home ownership andunemployment: does the Oswald thesis hold forAustralian regions?, mimeo, 2002.

6 If landlords sell as vacancy rates rise, moreover, theywill provide much of the housing supply for such newowners to buy.

7 26% of owner reference persons moved to be nearertheir job or because of a change in job, while 22% ofprivate renters did in 2001/2, according to the Surveyof English Housing.

8 ODPM Summary: Effects of divorce, remarriage,separation: new households.

9 Unless it induces people to lower their housingstandards by crowding up more, which is not what mostcommentators would advocate.

10 P. Balchin Housing Policy, 1995, p114.

27

IntroductionThe previous chapter looked at thecharacteristics of tenants. This one switches thefocus to the supply-side by looking at thecharacteristics of landlords and their motivesand also at the rise of agent professionals asimportant intermediaries between landlords andtenants. In addition, landlord-tenant relationsare explored and the debate over compulsorytenant deposit schemes is considered.

Having done this, the chapter focuses on somefinance and investment issues. The boom inbuy-to-let mortgages is examined, itssignificance in terms of the whole sector is putin context, and the future of mortgage lendingto landlords evaluated. This is followed byconsideration of the fashionable topic ofProperty Investment Funds (PIFs). Consultationover PIFs was announced in the 2004 Budget.They are planned to be similar to US real estateinvestment trusts (REITs), which now play asignificant part in that country’s real estatemarket, while versions of the approach havebeen developed in a number of other countries.Three interlinked PIF themes are studied: first,an examination of the general principles of thisinvestment approach; second, a study of recentexperience of similar investment vehicles inother countries, especially the USA and, third,an investigation of their potential roles in theUK’s privately rented sector.

The decline of the corporate and thegrowth of the individual landlord1

Private persons have grown considerably inimportance as landlords since the early 1990sand now are the owners of over two-thirds of alllettings; while companies (two-thirds of whichcurrently are specifically property companies)have rapidly declined in significance and now

represent only around 10% of all landlords.Other institutional landlords also seem to bewithdrawing with their letting share now wellbelow a fifth (Table 4.1).2 Many of these otherinstitutional landlords are non-commercialorganisations including governmentdepartments, educational establishments andcharitable trusts.

In part, the decline of the corporate landlord isassociated with the long-term shrinkage ofprivate renting, which has resulted in the sellingoff of estates by traditional large landlords. Thisaccelerated once rent regulations eased in thelate 1980s and was further encouraged by risinghouse prices from the mid-1990s onwards.

This shift away from the corporate sector iscontrary to government wishes, as it believesthat such landlords are more responsible, onaverage, and are more likely to keep apermanent presence in the sector. Whetherthese factors are actually true in practice is farfrom clear. The substantial withdrawal bycompanies from the sector, noted above, doesnot suggest any great stability in their holdings.

Unfortunately, there is actually little empiricalevidence on whether one type of landlord is inany way superior to another and any currentevidence might, in any case, be a poor indicatorof future behaviour as the sector is goingthrough such structural change. At present,agnosticism with regard to the issue of whether

Table 4.1 Types of landlord

% 1994 1998 2001 Private persons 47 61 65Partnerships 3 4 5Companies 25 22 13Others 25 14 17

CHAPTER FOUR

Ownership and investment

28

there is a preferable type of landlord is probablyin order, while recognising that there will be arange of efficiencies and competences whateverthe type of landlord, as in any other line ofbusiness.

Many landlords regard renting as a sideline totheir other activities. Only 15% of dwellings in2001 were owned by those who treated dwellingrenting as their main activity (includingcompanies as well as individuals). In contrast,there has been a big increase in recent years in‘side-line’ investors. So, those that focus theiractivities on residential letting are in a minority.In fact, only 40% of individual landlords derivemore than a quarter of their income from theindustry, two-thirds have other paidemployment and many others are retirees.

Companies and organisations tend to havemuch larger property holdings than individuallandlords: 60% of them own 25 or moreproperties, whereas only 8% of privateindividuals do. This means that the decrease incorporates and the increase in sideline privateinvestors has led to a fall in the average numberof dwellings owned by the typical landlord. Themedian number of properties for corporates andindividuals together in 2001 was only 4 (downfrom 9 in 1994). The range of holdings is large,furthermore, with 30% owning only oneproperty and, at the other end of the scale, justover a third having a portfolio of 10 or more.3

Four types of investorFour types of investor have been identified.4

• Business landlords These are propertycompanies or individuals who manageresidential property as a full-time job. Theyaccount for 15% of all privately renteddwellings.

• Sideline investors These are companies andindividuals not fully employed in the sectorand who regard their properties as aninvestment. They hold 45% of all dwellings.

• Sideline non-investors These areorganisations and individuals letting propertywithout regarding their holdings primarily asan investment. Like property companies, thesetypes of landlord are declining in importanceas the opportunity costs of a non-investmentattitude grow.

• Institutional landlords They do not viewtheir property holdings primarily as aninvestment and only get a minority of theirincome from it.

The last two types of non-investor together own40% of all privately rented dwellings. Thismeans that at most only three-fifths of landlordscurrently have what could be called a market-oriented attitude to their properties.

An industry with low economies of scaleThe growth of the small landlord illustrates afundamental characteristic of housingownership and management in general andprivate landlordism in particular. There appearto be no significant economies of scale beyonda very small number of dwellings, so that asmall landlord can be just as cost efficient as alarger one and may be even more so.5 Evenwhen they have no expertise, small landlordscan hire specialists as agents or rely on theadvice of solicitors and others.6

Financial factors do not indicate particular scaleadvantages either. The financial benefits of sizein the privately rented sector have probablyfallen with mortgage liberalisation. Largerlandlords, for example, no longer getpreferential mortgage terms. Instead, mortgagelenders now permit small landlords to signaltheir financial probity without imposing afinancial penalty simply because of their size.

In addition, the risk pooling benefits of owninga larger, dispersed housing stock are likely to besmall. Any landlord may end up buying a‘lemon’ of a property or by luck purchase an

29

exceptionally well-performing one. Yet, theinvestment pooling benefits that smooth outsuch potential variations end at quite smallscales of holdings. Larger pooling benefits fromgeographic diversification are also weak. Thevarious regions of the UK’s housing market overthe long run tend to move too closely in unisonfor a geographically spread out portfolio tooffer much return smoothing. Investor landlordswould be better advised to pool their propertyassets with other types of investment orinternationally.

The scale economies argument is an importantone for policy, as it suggests that there is nofundamental economic reason why theownership structure of the privately rentedsector should be of general policy concern.What exists over time may well change, as hasbeen observed in recent years, but that shouldbe regarded as indicative of the dynamism ofmarket forces rather than a signal forgovernment action. This, of course, is not to saythat governments should not encourageinnovative organisational forms, but these needto have specific justifications.

Are large landlords doomed?International evidence seems to point to theprivate rental sector being a haven for the smallinvestor. This does not mean there is not asignificant role for the larger investor, ratherthat there can be no expectation that large-scaleconsolidation will ever come to privately rentedsector investment.

There are several reasons why larger investorsare attracted and discouraged from investing inthe privately rented sector. Informationlimitations, regulation and uncertainty, plusreputation and management costs can act asdeterrents; whereas it has the traditionalattractions of being a relatively risk-free, long-term investment with a different cyclicalperformance from other asset classes.7 There isalso the opportunity to use expertise andacumen to spot changing market trends and

other profitable opportunities. In addition,larger investors are being currently boosted byfinancial innovations and the diffusion ofinnovations from other property sectors, such asin facilities management and in the provision ofa wide range of ancillary services and productpackages.8 There is also anecdotal evidence ofrenewed large investor interest in residentialproperty. The barriers to large investors,however, are still quite high, so thatmechanisms to reduce them are likely to lead togreater flows of capital from such sources intothe privately rented sector.

Landlord-tenant relationships Despite the cultural change in public attitudesto private landlords in recent years, a poorimage lingers in some quarters. The ghosts ofthe likes of the infamous Peter Rachman havenot been finally laid to rest. He helped to bringdown a government in the early 1960s andspawned salacious media coverage, reports andlegislation against landlord malpractices. Yet,such operators existed in an era when rentcontrols could bring sudden wealth when atenant quit – giving a huge incentive to somelandlords to misbehave. This history,consequently, should be remembered more forhighlighting the fact that regulation can channelbehaviour in unfortunate and unintended ways,rather than as illustration of any innatetemperament of landlords.

The juxtaposition of the ‘good’ and ‘bad’amongst landlords, none the less, is acommonplace of policy debate in a way in whichit is not for many other activities. Landlords,unsurprisingly, repost to such labels with thenotion of the ‘good’ and ‘bad’ tenant. A third oflandlords surveyed in 2001, in fact, had recentlysought possession on grounds of tenantbehaviour – and only half thought the processwas effective.

In part, the rush to categorise the behaviour oflandlords and tenants is because propertytransactions are complex and replete with

30

information deficiencies and asymmetries. Onthe one hand, landlords do not know for surewhether new tenants will pay the rent nor thatthey will treat the property well; on the otherhand, new tenants cannot properly assesslandlords’ true reservation rents nor whetherthey will respect fully the terms of the contract.

Furthermore, the landlord-tenant relationship isone where each can impose high costs on theother, which can lead to personal friction. Thereare high transactions costs if the relationshipcompletely breaks down: both for tenants inhaving to look for a new home and for landlordsin searching for a new, high quality, tenant. Thissituation leaves scope for continual bargainingand the potential for opportunistic behaviour.

Recognition of such potential landlord-tenantissues is not the equivalent of a red cardimplying inevitable market failure and the needfor a regulator to fix the problems. Instead,there are many market responses to suchpotential difficulties. An example is therequirement for deposits from tenants toencourage them to pay the rent in a timelymanner and look after the dwelling. In fact,more than three-fifths of market lettings requiredeposits and 70% references to improveknowledge about potential tenants.

In general, the potential for problems to arisebecause either party does not have fullinformation about the other acts as an incentivefor both landlords and tenants to get on witheach other. Such a mutually beneficial strategyminimises the stress, complications and costs ofdisputes. Such strategies are less likely to existin the more distant, bureaucratic and rule-bound settings common in social housing. Thismay help to account for the average higheresteem private tenants have for their landlordsthan do social tenants. Private landlords, forexample, score better in surveys of whethertenants are kept informed than do sociallandlords, and only 5% of private tenants, whenasked, are very dissatisfied with their landlordsin contrast to 11% of social tenants.9

Even so, such market-based solutions would beimproved if some transaction costs werelowered, especially the costs of acquiringinformation. Tenants, for example, are oftenignorant of their rights and either landlord ortenant may be unaware of the importance ofgood-natured negotiation. Landlords and theiragents repeatedly complain in surveys about thedifficulties of getting public information. Thisincludes such important matters as fire safety –one of the major concerns in current proposedlegislative change with regard to houses inmultiple occupancy.

Modern technology has not been used toimprove information as effectively as in manyother spheres. There is no privately rentedsector clearing-house web-site at either nationalor local government levels. Both the ODPM andthe Inland Revenue, for example, have web-based information but they deal only withselected matters that fall under their respectivejurisdictions.

Information is a public good and much of thenecessary information for both landlords andtenants is about public regulations and policies.This would consequently seem to be an areawhere public policy could significantly beimproved at limited cost with major benefit.

The problem of tenant deposit disputesThere has been widespread concern that somelandlords abuse tenant deposits by failing toreturn them or withholding all or part on thebasis of fictitious claims for damage. Similarly,it is reported that some tenants fail to pay thelast month or two’s rent, perhaps for fear of notgetting back their deposits or because they donot want to accept responsibility for damagethey might have caused.

Organisations such as the Association ofResidential Letting Agents (ARLA) and Shelterare arguing for the need for a public TenancyDeposit Scheme to hold deposits and

31

adjudicate over deposit disputes, but thegovernment in 2003 pulled out of initialproposals to set up one on the grounds thatvoluntary schemes were being initiated and, so,it was better to wait and see whether they workedbefore reviewing the situation. In response,ARLA is setting up its own voluntary TenancyDeposit Scheme for Regulated Agents, whichcame on stream in May, 2004. In it, landlordsand tenants agree to adjudication by a thirdparty over deposit disputes and an independentcompany has been set up to undertake that role.

The theoretical argument that it is in bothlandlords’ and tenants’ best interests to getalong, outlined in the previous section, isobviously weaker when the relationship is comingto an end. Mutual suspicion consequently mayeasily replace co-operation. So, it is unsurprisingthat deposit dispute issues are contentious.

In terms of policy options, the government’scurrent stance would seem to make sense. Thenotion of an independent adjudicator, whetherthey hold the deposits or not, is akin to aninsurance policy. Hopefully, it is not needed butit provides a cushion when required. When athird party scheme is well-run, moreover,potential abuses are lowered and overall marketefficiency is increased.

Most insurance, however, is costly to provideand tenant deposit adjudication and/or holdingthemselves similarly involve expenses. The costsof the system may be met simply by chargingthose that use the adjudication process or bycharging a flat, or risk-related, fee on all thatwant to use the scheme. If only those whoactually use the dispute process are chargedcosts, the charges imposed on them will be toohigh because there would then be free-riding ofall of those that want the implicit insurance ofan adjudication system but turn out not to needto make a claim. A mixed user cost and generalfee would be more efficient and equitable.

Should a deposit adjudication scheme becompulsory? In most markets, insurance is

voluntary; in some, it is compulsory. In theprivately rented sector, some tenants andlandlords will be more confident of a low risk ofdeposit disputes than others, so that the formerare less likely to want to pay the necessarydeposit-related insurance premium than theothers. A compulsory scheme would, however,impose costs on all landlords and tenants, sothat those that did not want the insurance wouldbe subsidising those that did. If the governmentprovided the scheme, they would either have tocharge user fees and/or impose a compulsorycharge on all – leading to the same problemsoutlined above – or the general taxpayer wouldbe subsidising landlords and tenants for nogood reason.

The only clear economic case for governmentintervention is when private insurance marketscannot be formed. This generally occurs whenbad risks crowd out a feasible market (adverseselection). Yet, this seems unlikely to be thecase with tenant deposit dispute schemes. Themodern private rental market is only 15 years oldand already private schemes are coming onstream. Following ARLA’s lead, otherindependent adjudicators might well soon setup and compete for business. The competitionwould be good for keeping the costs ofadjudication down. A compulsory public schemewould face no similar competitive pressures. Alltold, the government seems wise to wait.

The rise of professionalismAnother market-based solution to potentiallandlord-tenant problems is the existence ofprofessional lettings agents. For a fee, theymanage their properties and deal with tenantsand their problems on landlords’ behalf. Thisavoids the need for direct personal involvementby property investors; increases liquidity in localmarkets as investors no longer have to have alocal presence; and leads to a self-selectionprocess through which landlords who are leastinterested or adept at mastering the landlord-tenant relationship can opt out of it and hirean expert.

32

Many landlords now use an agent to managetheir properties. Just under 60% of dwellings‘accessible to the public’ were let and/ormanaged by agents in 2001 and the percentagemay well have risen since then. Dwellings letand managed by agents tend to have moreformalised management processes and are in abetter state of repair than others. As the marketrental sector is relatively young, it is to beexpected that the ‘demonstration effect’ of suchpractices on other lettings is likely to be highand, so, further improvements in lettings andmanagement processes are likely in the future.

As the market-based component of the privatelyrented sector has grown over the past fifteenyears, there has been a parallel blossoming ofthe number of agencies offering lettings andmanagement services. Many existing estateagents have set up or extended their lettingsdepartments and built up operations to assistlandlords in buy-to-let. Others have set up aspurely lettings agents. This has greatlyincreased market information in the sector bothfor landlords and tenants and, as a result,significantly improved the operation ofthe market.

The ease of entry and exit to the agencybusiness are high. Agency fees and services,therefore, are set in a competitive environment.This has encouraged the widespread adoptionby landlords of agency services. Continuingcompetitive pressures should both increase thelong-run quality of housing management andreduce its cost.

Not all lettings agents, of course, are efficientor honest and there are calls for compulsoryregistration. The case for this, however, doesnot seems strong and market forces are likely tobe more effective at weeding out the cowboys.

The buy-to-let boomCompetition amongst mortgage lenders wasargued in Chapter 2 to be one of the mainreasons why market-based renting has expanded

so significantly over the past 15 years. By thelate 1990s, mortgage lenders began to brandmortgage products specifically as ‘buy-to-let’mortgages in association with ARLA as a meansof encouraging additional individual investors.

The spectacular growth of buy-to-let can beseen in Figure 4.1. By the end of 2003, therewere over 400,000 buy-to-let mortgages – upfrom zero prior to 1998 when the scheme waslaunched – with an outstanding mortgage valueof £39bn. In addition, industry sources suggestthat another £10bn or so may be lent tolandlords through other mortgage instruments,giving a rough outstanding total loan figure tothe sector of £50bn.

Not all buy-to-let mortgages represent newrental dwellings coming onto the market.Though newly issued buy-to-let mortgages areoften associated with extra purchases of rentaldwellings, many are re-mortgages of existingproperties to take advantage of improved termsand better interest rates. This may help toexplain the paradox whereby buy-to-let dataseem to suggest a significant expansion of themarket-rental sector in recent years, whereasgovernment surveys do not.

The growth in buy-to-let also has to be put incontext of the overall size of the privatelyrented sector. Total buy-to-let mortgagesrepresent less than a fifth of privately renteddwellings as a whole and less than a quarter ofthe market-sector specifically. Back of theenvelope calculations suggest that under a thirdof the privately rented sector stock is mortgagedto any significant degree. These calculations areconfirmed by the English House ConditionSurvey, which reported that in 2001 only 31% ofprivately rented sector dwellings weremortgaged. Even if all these mortgages wereconcentrated in the active investor sectorsidentified above, it would represent only half oftheir holdings. Whichever way the numbers arelooked at, consequently, it would seem that theprivately rented sector is currently under- ratherthan over-mortgaged.

33

A lack of financial sophistication apparentLow sector indebtedness ratios suggest thatmany landlords are not taking advantage of thebenefits of financial gearing. Mortgageborrowing can raise investor returns as long astheir total rental returns are greater than thecost of borrowing net of tax. In such situationslandlords can leverage the returns on their owncapital significantly by taking out mortgages upto prudent levels that shield them againstmortgage default risk.10 Many landlords,surprisingly, fail to do this; even thoughmortgage interest payments are tax deductible,meaning that the net-of-tax cost of borrowing isparticularly attractive.

The low gearing characteristic of the privatelyrented sector suggests that:

• Many private landlords are still relativelyfinancially unsophisticated

• Mortgage lending to private landlords willgrow significantly in the future

• The private rental sector could operate onlower gross returns, if potential gearing andassociated tax effects were adopted moreeffectively.

Will buy-to-let cause a housing market crash?The expansion of the buy-to-let market hasaccelerated in recent years and led somecommentators to predict a bubble-style crash asa result in the near future on the grounds ofwhat goes up must come down. On this doom-laden scenario, problems will spill over into thehousing market as a whole and be the mainreason why house prices rapidly fall by 20 to50% in nominal terms depending on the sourcesin order to realign themselves back onto theirlong-run ratio with average earnings.

How many buy-to-let investors are actuallyteetering on a financial precipice is disputed.While some commentators suggest the numbermay be high, mortgage industry representativespoint to the strict lending criteria that are inplace. In 2003, the maximum loan-to-value ratiowas 80% and the minimum rental cover ofmortgage repayments was 130%, giving asignificant cushion before financial oblivion isreached. The probability of a buy-to-let crash,moreover, is diminished when the privatelyrented sector is viewed overall. Rather thanbeing over-borrowed, it is under-borrowed, asexplained above.

One version of a crash scenario is that it wouldbe triggered by an unexpected rise in the rentalvacancy rate, caused by supply outstripping

34

0

350,000

400,000New advances

Number of mortgages

Outstanding numbers

300,000

250,000

200,000

150,000

100,000

50,000

450,000

1998 1999 2000 2001 2002 2003

Figure 4.1 The Buy-to-let Boom

Source Council of Mortgage Lenders

demand. Why such large increases in vacanciesshould take place is a mystery in a situation ofgeneral housing shortages. Even in the unlikelyevent of a sudden surge in vacancies, it isunclear why the impact should be concentratedon the relatively small heavily-mortgaged part ofthe privately rented sector.

Current increases in rental supply, as shown inChapter 2, do not, in any case, seem that great.Yet rapidly expanding supply, rather thancollapsing demand, is likely to be a pre-requisite of any marked rise in vacancies. Onthe available data, therefore, over-supplyproblems in the privately rented sector, even ifthey were to occur, would most likely belocalised rather than national events.

Another crash scenario bases itself on marketsentiment. As soon as overoptimistic buy-to-letinvestors see that house prices are no longerrising rapidly, they will try to sell off theirholdings and, in doing so, accelerate the overallhousing market into a slump. For a generalslump to occur, there has to be a significantexcess of sellers in both the rental and owneroccupier markets that generates a large numberof vacancies. This is again unlikely in a nationfacing marked housing shortages. If sales ofrental properties arose, for example, shortageswould grow in private renting pushing up rentsand, therefore, returns. Selling landlords couldalso probably reduce their dwelling holdings byselling out to new or existing owner occupiers,even with a slight softening of prices. Activity inrental and home owner markets tends to movein different directions over the course of thehousing market cycle as households delay orspeed up their planned entry to homeownershipdepending on market conditions.

An issue here is whether new buy-to-letlandlords are in the business of home rentingfor the long haul or simply for short-termcapital gains. There is obviously no clear way ofknowing precisely. Yet the expectation must bethat many are in for the long-term. Many, forexample, claim that private rental investment is

part of a long-term strategy to build up apension nest egg. In a recent ARLA survey inFebruary 2004, more than 90% of landlordrespondents said that they would not sell ifhouse prices fell and nearly two-thirds expectedto hold their properties for at least ten years.Housing transactions have high costs, whichdeter strategies of rapid entry and exit.Landlord investors have the near certainprospect of long-run rises in house prices, sothey would jeopardise long-term gains if theysold up in a falling market.

Although forecasting is an uncertain business,the prospect, therefore, must be that marketswill work in the sense of bringing aboutadjustments that do not involve a temporary andbloody market collapse. If disequilibrium didoccur, supply is more likely to be brought backinto balance with demand by some gradualstock-adjustment mechanism rather thanthrough a cataclysmic crash process induced bydefaulting buy-to-let landlords.

Property Investment Funds (PIFs)Earlier, it was pointed out that there are somesignificant barriers to investment by largeorganisations in the privately rented sector.PIFs are the most recent proposal to improvethe situation and the government is currentlycontemplating their introduction. This sectionevaluates the potential benefits and limitationsof such an innovation. Part of what follows issomewhat more technical than the rest, as itoutlines the general principles and operation ofa particular type of property investmentvehicle, so some readers may wish to omit itand go directly to the discussion of residentialPIFs.

BackgroundTax-efficiency is the central feature of PropertyInvestment Funds (PIFs) and it is hoped thatthey will encourage investment in residentialproperty. Yet, because they are going to begeneral property investment vehicles, it is usefulto explain their broad features first, before

35

examining their potential roles in the UK’sprivately rented sector specifically.

PIFS do not exist in the UK at present. The UKproperty industry has been lobbying for anumber of years for the introduction of a USReal Estate Investment Trust (REIT) style vehiclefor all classes of property. After several rebuffs,the Treasury in documents associated with theNovember 2003 Budget Statement suggestedthat they might finally be introduced soon and aconsultation period followed the March 2004Budget.

REITs originated in the United States in the1960s, since when the name REIT has becomethe generic term internationally. They have alsobeen introduced under a variety ofnomenclatures in a range of countries, includingthe Netherlands, Belgium, France, Turkey,Japan, Singapore, Hong Kong, Taiwan andAustralia. They are also currently beingproposed for a number of other countries,including Germany, leaving the UK increasinglythe odd one out.

After all, tax haven property companies – andeven foreign-based REITs in principle – caninvest in the UK, so that there is a substantialthreat that indirect property investment vehicleswill increasingly quit the UK for elsewhere. Thatwould both threaten highly-paid British jobsand rob the Treasury of tax income.

The decline in UK residential corporateactivity, noted earlier, which in value in theproperty world is relatively small change, hasbeen mirrored in much higher valuecommercial property, where institutionalholdings in property relative to other assetclasses have fallen substantially over the pasttwo decades11 at the same time as propertyfunds have gone off-shore to limit taxexposure. So, there is a growing need toreverse the dramatic relative UK decline ininvestment in property of all types as an assetclass over the past decade or more.

Global pressure apart, REITs are clearly afashionable financial vehicle; one that mayincrease liquidity and investment in UK privaterental housing, as is explained below.

OperationThe rules for REITs vary between countries, butthe principles are common and straightforward.REITs issue shares and borrow under clearlydefined regulatory formulae. They actpredominantly as pure low-risk propertyinvestment vehicles, distributing the majority oftheir net rental income to investors underadvantageous tax rules. They tend to specialisein one type of property class, such as residential,mobile homes, retail, offices and so on.

With regard to ownership and activities, a REITowns and manages a stock of property assets (or,more usually has them managed on its behalf).The UK property industry12 proposes a closed-end REIT structure, with a fixed number of sharessold at one point in time, after which the REITs‘shares trade on a secondary market, typically theStock Exchange. Ownership is passive in thatREITs usually are not permitted to undertake themore risky activities associated with property,including new development and refurbishmentor extensive dealing in property assets throughan active policy of purchases and sales.13

There are country-specific rules on how manyshares a REIT may issue and what proportion ofassets can be funded through borrowing. Forexample, in Belgium, gearing is limited to 50%;yet in France it is unlimited. Typically, theregulatory framework for REITs imposes gearing(leverage) ratios that are low by property marketstandards. A common suggestion in the UK, forexample, is that borrowing should be capped atonly 50% of a PIF’s assets, with the restfinanced by equity.

The net income from REIT holdings of propertyassets are mainly passed through directly toshareholders as dividends (a minimum of 90%under US rules; somewhat different shareselsewhere). REIT profits are not subject to

36

corporation tax, but shareholders pay tax ontheir dividend income and on any capital gainsthey have from increases in REIT share prices.14

In the UK, they would also pay 0.25% stampduty on any share transactions they undertake atcurrently prevailing rates.

The benefits of the REIT modelThere are several suggested benefits of REITs:

1 Tax efficient vehicle for indirect property investment

As REITs are not subject to corporation tax,unlike traditional property companies, andshareholders receive higher proportions ofrental income from property as dividendsbenefiting REIT shareholders over many otherindirect investment possibilities. Currently,much indirect property investment (such aspurchasing the shares of a property company) issubject to double taxation in comparison todirect ownership, which has encouraged thedecline of the indirect form of property holdingin the UK. The REIT vehicle by removing thisdouble-taxation effect may encourage moreinvestors to own property indirectly through aUK PIF’s shares rather than to hold it directly ina non-corporate form (for example, as a directinvestor or an owner-occupier), as an institution(pension and other funds pay no tax on theirincome derived directly from property) or via anoff-shore fund.

The removal of the double-taxation effect withREITs does not necessarily mean that thegovernment loses out in tax revenues. Instead itis generally argued that the direct tax effect ofthe changeover is limited or may be evenpositive for Exchequer revenues. This mayoccur, for example, because any capital gains onREIT shares are more transparent andimmediate than are accounting revaluations ofproperty holdings in a property company’sportfolio of assets or the realisations of capitalgains by individual property investors.

The tax take could also be significant on a one-off basis, if large amounts of existing property

were put into REIT structures and a conversiontax was levied. For example, the WellcomeTrust, the medical charity, is contemplatingconverting its large holdings of residentialproperties in up-market South Kensington inLondon to a PIF, when permitted. £200m ofproperties would enter the Wellcome PIF, givingthe Trust a huge cash windfall of which thegovernment might well insist on having a shareas a conversion or windfall tax.

For all REITs, moreover, it is argued that theREIT vehicle encourages investor interest inproperty share holding and dealing, so that thestamp duty income from share purchase andtransactions on the secondary market help tooffset any potential losses of direct tax revenue.

Such arguments obviously depend on estimatesof structural changes in property marketinvestment behaviour and holdings. The InlandRevenue has traditionally been wary,emphasising the potential tax losses of awholesale switch of property assets to the REITformulation and the privileged status propertywould have over other asset classes, with thelatter often not having equivalent tax efficientinvestment vehicles. General and governmentopinion seems to be moving against such apessimistic perspective.

2 Increase market liquidity and efficiencyProperty is an illiquid asset. It takes both timeto buy and sell property directly andtransactions costs are high. This means thatinvestors cannot alter their exposures in atimely manner to changed market or personalcircumstances.

Owning the shares of traded REITs partlyremoves these problems as their shares are easyand cheap to buy and sell. In so far as REITslower the transaction costs and speed up theoverall time of changing investment positions,both the scale of investment in property and itsliquidity should be increased. This should raisethe efficiency of property markets.

37

Being such a transmission mechanism, however,implies that REIT shares contain a degree ofvolatility, because of the mismatch between theinvestment characteristics of their traded sharesand of the underlying assets they own. A REIT’sown investments, after all, are direct propertyinvestment. So, they cannot rapidly alter theirown investment positions at limited cost incontrast to the holders of their shares. Thisimbalance is likely, in principle, to lead to ahigher degree of volatility in their share pricesthan that of returns in the direct market.15

Experience in other countries has varied, partlybecause of the time horizons over which REITshares have been traded and partly because ofthe rules and regulations associated with REITstocks. In some countries, such as Belgium,REIT performance has tended to be in line withdirect property. In Australia, in particular,(where they are called Listed Property Trusts,LPTs) LPT shares have become hot and heavilytraded stocks in recent years. In the USA,experience has been different with a hybridperformance of REITs that has been somewherein between the volatilities of direct property andequities.

3 Lowering the lumpiness of propertyinvestment

Direct property investment is the acquisition ofa lumpy asset in that whole buildings or, withresidential, dwellings have to be purchased. Thefunds required may be too much for many smallinvestors and with larger and more expensivestructures, even for many major investors. Thescale of any allocation of funds required,furthermore, may skew investors’ overallholdings away from optimal portfolio mixes.Lumpiness, therefore, discourages manyinvestors from direct property investment.

In so far as REITs extend the range anddivisibility of indirect property investments, theymay reduce lumpiness effects in propertyinvestment. In an ideal world, REITs act likewholesalers do in any market: expertlypurchasing large bundles of assets and then

breaking down return outcomes into share-unitportions as quasi-pass-through vehicles forinstitutional and individual investors.

The crux of the REIT argument, therefore, is thatthey improve liquidity and enable a wider array ofinvestors to enter property markets. Thisbenefits property markets and investors whohave property, as an asset class, to add easily andcheaply to their portfolios. The improvement inliquidity for property markets means that theybenefit from a greater flow of funds and morecompetitive pressures. This increases their scopeand depth and, hence, their efficiency.Nonetheless, the investor impact is a particularlyimportant REIT benefit as property investment istypically a safe, low yielding asset andfluctuations in property returns are only weaklycorrelated with those of other financial assets.

The above characteristics could be said to betrue of many indirect forms of propertyinvestment relative to direct investment, but theREIT approach is argued to accentuate thepositive and disregard several negative featuresof indirect property investment (usuallyassociated with the risk of investing in propertydevelopment that REITs are not allowed toundertake). This means, so it is argued, thatREIT property values have a lower tendency totrade at less than underlying net asset value(NAV) than is the case with other indirectproperty investment vehicles. Consequently,REITs are said to have a particularly effectiveimpact on property efficiency.

4 TransparencyIn order to protect share values and, thus, theirjobs and bonuses, REIT managements have astrong incentive to be as transparent as possibleabout property asset holdings and performance.In order to achieve this, REITs tend toundertake frequent valuations of their realestate assets. Moreover, they tend to invest onlyin prime property and standardised assetranges. Transparency needs also help to explainwhy limited numbers of REITs invest in morethan one property class, even when they are

38

permitted to by legislation. In the USA, forexample, the main residential REITs areseparate entities from retail and office ones.

Yet, REITs can still gain bad reputationsthrough overly-risky investments or other formsof under-performance (as they did in the USAprior to the late 1980s). So, they can alsosuffer problems and find their shares trading athuge discounts to NAV. The ratio of shareprice to NAV, consequently, is also aboutprudent management and sensible andtransparent asset holdings as well as the taxefficiency of the REIT structure. This highlightsthe importance of setting up a REIT regulatoryframework that creates the appropriateincentive structures.

Overseas experienceThe REIT structure helps to explain the runawaysuccess of such property ownership vehicles inthe countries where they have existed over thepast decade. This is especially the case inAustralia and the USA.

The growth of REITs in number and value in theUSA since the early 1970s is shown in Figure4.2. The spectacular rise in their capitalisationvalue over the past decade is clearly apparent inthe diagram, though mergers and acquisitionshave reduced their number by about a quartersince the mid-1990s peak.

The expansion of US REITs had much to dowith reforms to the country’s tax laws in the1980s, which made other forms of propertyownership less attractive; favourable taxconditions when converting existing real estateholdings into the REIT format; and in propertymarket conditions at the time. These factors incombination turned REITs into a growth stockin the 1990s, raising market expectations ofrapid capital gains which pushed REITcapitalisations above underlying asset values.

Overvaluation, combined with the 1998 Asianfinancial crisis and the hi-tech stock marketboom, temporarily took the shine off the REIT

market. But, more recently, a benign interestrate environment and the switch of investorinterest to property following the bursting ofthe hi-tech bubble has revived REIT shares andturned them into a growth stock once again.16

Recent experience in the USA shows that theREIT market there has been subject to notablevolatility, arising from swings in marketcircumstances and investor sentiment, though itshould be noted that more of the fluctuationshave been borne in share valuations than inunderlying REIT property asset holdings.

To an extent, it can be concluded that recentREIT growth in the USA and other countries hasbeen cyclical and associated with variations overtime in the expected total returns of particulartypes of asset. Yet, the new scale of REITs in USinvestors’ portfolios indicates that there hasalso been a permanent change in asset classholdings. The modern US REIT market, beingonly somewhat over a decade old, is still toorecent to be able to make firm predictionsabout its long-term behaviour.

The limitations of REITsThe potential benefits of REITs should not beunder-estimated in improving property marketfunctioning through increasing liquidity,transparency and competition for investorfunds. Nonetheless, there are drawbacks to theREIT framework. The principal difficulties, asfar as the property market is concerned, relateto the rigidity of the REIT structure. This meansthat it is not appropriate for all marketcircumstances and organisations.

The fact that REITs specialise in a narrow rangeof property types, for example, leaves themexposed when the market in those assets turnsdown, with no cushion available from a morediversified asset structure. Within any propertytype, furthermore, to sustain shareholderinterest and ensure transparency of assetholdings, prime, modern, standardisedproperties are likely to prevail in REITportfolios, which again narrows the potential

39

benefits of portfolio diversification. This helpsto explain potential REIT share volatility.

The requirement inherent in the notion of‘passive’ investment to separate developmentfrom asset ownership can also limit the propertyreturns available to investors. More risky, buthigher return, property activities cannot beundertaken by REITs. Like their underlyingassets, therefore, they are a low risk, low returntype of stock over the long-term that maydisappoint many hoping to see substantial long-term financial gains from the introduction of theREIT model into the UK.

REITs by being passive investment vehicles havelimited opportunities to influence the revenueearning characteristics of the properties theyown. Upgrading property portfolios, orchanging the mix of uses in a development toreflect changing patterns of demand,overstretches the notion of passivemanagement, because such actions are clearlyactive portfolio management.

The UPREIT and DOWNREIT formats wereintroduced in the USA in order to make itpossible for REITs to work with partners in thedevelopment pipeline. Even so, in the USA,REITs are always chafing at the constraintsimposed on their activities, which stronglyrequire them to be passive property investors in

order to qualify as a quasi-pass-through entitythat should not be subject to corporation tax.

Residential REITs Of particular concern to the privately rentedsector, of course, is the development ofresidential PIFs. The general arguments forREITs outlined above in terms of marketliquidity and efficiency hold for residential asmuch as for any other property sector and sodo not need to be considered again, though itshould be emphasised that they are asimportant in the residential sector as in otherclasses of property. Even so, three specificissues do take on particular importance in theresidential sector. These are:

1 What type of market segment wouldresidential PIFs invest in and with whatresults?

2 If PIFs were successful, what would be theeffects on the size of the privately rentedsector and its relationship to other housingtenures?

3 Can any special cases for PIFs be justified inthe housing sphere?

40

0

50

100

150

200

250

1971

1972

1973

1974

1975

1976

1977

1978

1979

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

0

50

100

150

200

250

$bn

number LHSmarket capitalisation RHS

Figure 4.2 US REITs number and capitalisation 1971-2003

Source NAREIT

Residential PIFs in the current housingenvironmentThe experience of the USA, the most mature ofthe REIT markets, is illustrative of the likelyimpact of PIFs in the UK privately rented sector.This can be studied by looking at the size andcharacteristics of the 21, stock exchange quoted,US residential REITs.17

The total market capitalisation of suchresidential REITs in the USA is $34 billion.Though a significant sum, this is less than oneper cent of the total value of the US privatelyrented housing stock.18 This obviously impliesthat residential REITs have had a limited impacton the privately rented sector in the USA.Small-scale private landlords still predominate,as they do here and in virtually all countries’privately rented sectors.

Yet, it could be argued that the impact of USResidential REITs has still been substantial insome localities, because REIT activity isconcentrated in certain growth regions.Furthermore, the sector has only really beenactive for a little over a decade, which inproperty terms is still early days. The impactmay consequently be greater over the long-term.

Other characteristics of US residential REITsare illustrative of what might happen here. Theyspecialise in the middle-to-upper parts of theprivately rented sector and invest in large-blocksof new property. They often invest in partner-developer built large suburban apartmentcommunities (via the UPREIT format). Olderproperties and piecemeal property holdingsimply high transactions costs and a lack oftransparency in asset holdings and so do notform part of US residential REIT portfolios.

There are interesting size characteristics aswell. The largest US residential REITs have amajor share of the total market capitalisation ofthe sector. This can be seen in Figure 4.3,where US REITs are listed sequentially bymarket capitalisation. The largest onerepresents almost a quarter of the sector’s total

market capitalisation and the top three aloneconstitute half of it. This suggests that there areconsiderable economies of scale for residentialREITs in contrast to other types of landlord,probably associated with their ability to raise capital.

The US experience, consequently, does notsuggest that a dramatic transformation of theprivately rented sector is going to happen as aresult of the introduction of PIFs into the UK,particularly in the short-run. This also meansthat the impact on other housing tenures islikely to be small. In addition, in the absence ofother incentives, they may operate primarily inan area of the privately rented sector that isalready well-served by small landlords – thoughhow direct the competition would be is unclearas REITs could not feasibly invest in the olderstock that individual landlords typicallyspecialise in.

Caution on the limited impact of PIFs is alsosignalled by the earlier failure of HousingInvestment Trusts (HITs). These were asecuritised vehicle introduced in 1996 that wereto offer lower corporation and capital gains taxin order to encourage City institutions to investin residential property. Yet no HIT has beenlaunched – partly because of a reluctance toinvest in residential and because of concomitanttax changes that weakened the attractiveness ofthe HIT model.19 However, the HIT vehicle is notfully tax-transparent unlike PIFs. There is alsothe benefit of a broader PIF market than wasever likely for HITs as other forms ofcommercial property are likely to be packagedinto the PIF framework as well as housing.

This somewhat downbeat view derived from theUS residential REIT and UK HIT experiencedoes not mean that there are no potentialbenefits to be derived from the REIT/PIF modelin the UK. Two other features of the USexperience are apposite here. Namely, theirfocus on new properties and the ability of someREITs to mobilise large sums of capital into theprovision of privately rented housing.

41

The Interim Barker Report identified areluctance of housing developers to build highdensity, high rise private developments.20

The PIF approach could, therefore, help toovercome this reluctance and generateadditional new housing output for the privatelyrented sector. This would benefit housingaffordability overall by increasing housingsupply in areas of high demand. PIFs may alsobuild such structures in niche rental markets,such as that for student housing. As thedemand for private renting tends to beespecially strong in inner city locations,furthermore, such built forms would encouragebrownfield housing development/redevelopment– another target of current government policy.

Residential PIFs, more generally, could play animportant role in urban regeneration by buildingapartment blocks for renters in such areas. Inlocations, such as the Thames Gateway area tothe east of London, the PIF format could have asubstantial impact and greatly increase thelikelihood of a significant number of privatelyrented properties being made available.

An issue arises in urban regeneration areas,however, in that building rental apartments inthose localities may be of a higher risk than isacceptable to typical REIT investors. This may

mean that for the residential PIF model tooperate in such contexts their returns mighthave to be higher to compensate for theadditional risk and that some form of incomeguarantee or building subsidy may be necessary.It may well be the case, however, that the PIFformat is a good way for public subsidies to beused to leverage large amounts of privateinvestment into socially desirable housingschemes.

Expanding the privately rented sector into socialhousing spheresResidential PIFs could also be used to acquirenewly built privately rented housing in marketareas more traditionally associated with socialhousing, including some of the more recentinitiatives for ‘cost-rented’ housing provision.Similarly, some existing social housing – muchof which has high refurbishment costs associatedwith it – could be improved and sold off to REITsfor future private renting. Belgian experienceoffers a useful precedent, because recently theBelgian government announced plans to set up aspecialised REIT to provide funds and expertiseto bring the country’s dilapidated social housingstock up to modern standards.21

REITs would also have another advantage overmore traditional forms of social housing and the

42

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

$m

Figure 4.3 US Residential REITs by market capitalisation 2004*

*Numbers on the x-axis list the largest quoted residential REIT as 1 through to the smallest as 21.Source NAREIT

private finance initiatives associated with it.They would provide more transparent publicinformation in terms of costs and returns,because investors in PIF shares would demandsuch information to be public knowledge. Thisinformation would then be useful for bench-marking social housing institutions’performance.

It is probably unwise, nevertheless, to use theREIT format to perpetuate non-market rents orheavily housing benefit subsidised rents,because that would make such REIT/PIFinvestments highly dependent on thecontinuation of such subsidies.

1 Data in this section are drawn from the English HouseCondition Survey’s Landlords’ Survey 2001/2,ODPM.

2 Extrapolating to 2004 from the data in Table 4.1.

3 The value of holdings would be a better indicator ofinvestor commitment than the number of propertiesowned, because one expensive property could easily beworth several cheaper ones.

4 Crook, ADH and PA Kemp (1996) ‘The revival of privaterented housing in Britain’. Housing Studies. The latestdata using these categories are in the English HouseCondition Survey’s Landlords’ Survey 2001/2,ODPM.

5 This conclusion is based on the fact that small landlordshave always predominated in any country’s privatelyrented sector. Actual detailed cost studies of economiesof scale unfortunately do not exist. One reason for lowscale economies is that large organisations can suffer adivision of interest between principals (property owners)and agents (managers), may develop inflexiblebureaucratic structures (which is particularly problematicin the shifting terrain of housing provision) and othertypes of inefficiency (‘cultures’, etc. – often called X-inefficiencies). These characteristics, for example, helpto explain inefficiencies amongst larger social landlords.

6 Small landlords prepared to undertake all managementand repair tasks themselves obviously pay no VAT onthose services, unlike those that use professionals.

7 These issues are discussed in greater detail in Crook,ADH and PA Kemp (1999) ‘Financial institutions andprivate rented housing’. York, York Publishing Servicesfor the Joseph Rowntree Foundation.

8 Some international evidence of such changes is reportedin The Private Rented Sector and Institutional

Investment: Lessons From Overseas A.D.H.Crook,Joseph Rowntree Foundation, 2000.

9 Survey of English Housing 2001/2.

10 Mortgages lenders, as noted earlier, currently treat thislevel as being met by a combination of loan-to-propertyvalue and net-rental-income to mortgage outgoingsratios.

11 Pension funds, for example, had 15% of their portfoliosin commercial property in the early 1980s and only 4%in the late 1990s.

12 See the Investment Property Forum website:www.ipf.org.uk.

13 In the USA, they are allowed to go into partnership withothers that undertake such activities in the UPREIT andDOWNREIT formats.

14 Conversely, they can write off any losses against theircapital gains on other investments – though, onceagain, the precise rules vary from country to country.

15 See Ball, M., Lizieri, C., Macgregor, B. TheEconomics of Commercial Property Markets wherethere is also an extensive theoretical and empiricaldiscussion of property investment and finance issues.

16 Thanks to Peter Brady for some helpful information onUS REITs. See Brady, Peter J. 1998. “A Look at theResurgence of REITs.” Mimeo. Federal Reserve Board ofGovernors, Washington, DC. January 1998. Noresponsibility for any factual errors or commentary onthe US situation here lies with him.

17 Source: NAREIT.

18 A UPE Consultancy estimate.

19 ADH Crook et al. 2002 Investment returns in theprivate rented housing sector, British PropertyFederation, London.

20Review of Housing Supply. Interim report –Analysis. K. Barker, UK Treasury, London, 2003.

21 See the Urban Land Institute’s web site: www.uli.org.

43

IntroductionConcerns about measured standards of housingquality and safety in the privately rented sectorhave grown over the past decade. A significantsegment of poor condition housing is argued toexist. The consensus view, without doubt, isthat all houses should be fully repaired andmeet strict officially laid down quality and safetycriteria. To achieve this, action is neededbeyond the requirements of existing legislationon repairs and standards that can currently beenforced through the courts.1

Though the government has resisted the moredraconian arguments for regulatory measures todate, it has still accepted some, which are nowpassing through Parliament in the currentHousing Bill. Claims about standards in theprivately rented sector, therefore, are leading tore-regulation in the privately rented sector. Is abrief era of a loosening of government controlover private renting coming to an end?

The most extensive area of new compulsionrelates to the mandatory licensing of houses inmultiple occupation (HMOs). The ones now tobe licensed in England are multiple dwellingstructures of three or more storeys containingfive or more residents. Compulsory HMOlicensing has already been introduced inScotland and Northern Ireland and has recentlybecome law in England.

These pieces of legislation criminalise anylandlord that does not register with the localauthority as a landlord of an HMO andundertake the appropriate actions associatedwith being registered. Such actions will include,amongst other things, bringing dwellings up tothe appropriate defined standards,demonstrating capabilities in housing

management and funding, and playing a part intackling the anti-social behaviour of tenants.There will also be compulsory registration forlandlords in areas of low housing demand – asdefined by the local authority – and for other‘selective groups’. In Scotland, licensing hasbeen associated with compulsory fees imposedon landlords to fund the scheme, which hascaused unprecedented resentment amongst thelandlord community.

These requirements, furthermore, are beingimposed on the privately rented sector alone, asthe legislation applies neither to social housingnor to owner occupation. If a group of studentsrented separate tenancy bed-sit rooms withsome common facilities in a three storeybuilding from a landlord that structure wouldthus be subject to the licensing rules. If theybought an equivalent house on a joint basis andlived in it in the same way as joint homeowners,licensing would not be an issue. Yet, the livingcircumstances and the risks associated withthem are presumably similar. The legislation isconsequently discriminatory, because theprivately rented sector is identified specificallyas a problem area. The response fromsupporters of the current legislation is thatpeople need more protection from the actionsof private landlords and other private tenantsthan they do from other members of their ownhouseholds. The exclusion of social housing ispresumably on the basis that social landlordsare more responsible and better able to dealwith potential inter-tenant difficulties. Boththese arguments may seem contentious to many,so that the discrimination charge still has merit.

It is not the purpose of this chapter to gothrough every aspect of recent attempts at there-regulation of the privately rented sector.

CHAPTER FIVE

Quality regulation: The re-introductionof Government control

44

Instead, this chapter will question a solereliance on regulation to improve the privatelyrented sector and also the use of external‘engineering’ standards as the only true theoryand identifier of the optimal level of housingquality. It will also highlight some unintendedconsequences that may arise as a result ofadopting such strategies.

A fundamental point is being made. Contrary towhat had been literally centuries ofinterventionist ascendancy, most housinglobbies and commentators have over the pastdecade embraced free market mechanisms asbeing the only means of ensuring that theprivately rented sector will survive as a viableand vibrant part of the UK housing scene.2 Theystill seem to accept that principle broadly withrespect to the price of rental accommodation.Yet many housing commentators, and thegovernment, are now beginning to reject freemarket principles in terms of the setting ofhousing standards or quality. Housing quality ispart of the quantity of rental housing provided,in that a lower quality means that a smalleroverall quantity of housing services is beingprovided by a specific dwelling unit. Controllingquantities in a market is as much an interferencewith free market processes as controlling pricesand should, therefore, not be done unlessabsolutely necessary.

The issue of quantity regulation is complex,because there are several means by which it canbe conducted and the empirical effects of eachof them can be interpreted in several ways.Regulation includes both legislation and thedegree to which it enforced in practice. As thelaws clock up in number, it may seemparadoxical that the number of enforcementorders carried out on private landlords hasactually fallen in recent years. This may suggestless regulation in practice, a softer touch, or itcould indicate the opposite, a harder touch withmore effective enforcement of the law, whichdiscourages the undesired behaviour and,therefore, leads to a lower incidence of theevent. Alternatively, less actual enforcement

could result from factors entirely unrelated tothe legislation itself. The growth of a free rentalmarket over the past years, for example, hasbeen associated with more tenants on averagewho can afford higher rents and want goodquality accommodation at the same time asgreater competition between landlords hasencouraged them to provide that mix.

Regulation may also have weaker forms thandirect coercion and criminalisation. Forinstance, discretionary improvement grants havebeen switched from owner occupation to theprivately rented sector. This is an incentive formof regulation, because the promise of cash toraise their rental incomes and lower their costsencourages landlords to behave in the waysrequired by the gatekeepers of thosediscretionary grants.

Whatever the means of regulation, the point isthe same. The various tiers of government wishto steer private landlords into deliveringhousing of a particular type quality-wise. Thetwo key issues are: 1. are the directions the rightones, taking into account any indirect orsecond-round effects that may occur? 2. Even ifthe outcomes are desirable, are the powers andthe implementing agencies any good atachieving them? Principle and practice areintertwined and policy may fail on either count.

The issues focused on here relate to housingconditions and the thrust of the ‘Housing Bill’legislation that recently passed its third readingin the Commons. Arguments such as the onesbeing made here are often misunderstood asbeing hard-hearted, because concerns abouthousing conditions in the privately rentedsector primarily relate to health and safetyissues. People should never be exposed tohealth and safety risks, many would believe. It isimportant to stress that it is not the intention ofthe argument being made here that peopleshould be exposed to unreasonable risks, ratherthat the heart has to be led by the head in apolicy context.

45

The argument is that risk and health preventionmeasures have costs as well as benefits and thatthe costs can easily outweigh the benefits. Nofull cost-benefit calculation to the author’sknowledge has ever been undertaken whensetting out house safety or condition measuresnor in setting minimum housing requirements.The world is full of risks. It costs resources andcloses off consumption possibilities to limitthem, which are reasons why no-one opts for azero risk life. Why should housing be anydifferent?

Furthermore, the mere existence of regulationdoes not make problems go away. That obviousfact can be illustrated simply by noting thatcrimes still continue despite extensive laws andcrime prevention measures. Why shouldhousing regulation necessarily be effective? It,after all, has not been an unqualified success inthe past.

Safety and housingSafety issues with housing are important. Therewere around 4,000 deaths related to homeplace events in 1998, for example, more thaneither traffic or work-related ones.3 Improvingsafety is clearly a good idea, but the benefitshave to outweigh the costs. If this decision ruleis used, the case for additional legislation maybe strong. There is a worrying concern,however, that it is not being adopted in muchpolicy debate over the privately rented sector.

In calculating the likely cost-benefit ratio ofsafety regulation, four factors have to bebrought into consideration that do not seem tooccur at present.

1 In order to calculate the benefits, a theory ofcause and effect must be proposed, empiricaloutcomes estimated and an aggregate valueput on the lives saved or injuries avoided byundertaking specific actions. Both the theoryand results need to be subject to scrutiny.Simple broad-brush correlations are notenough, because that might mis-specify

causes and lead to wasted actions andexpense. Yet, such correlations seem to bethe basis for much current housing qualityand safety commentary. The case forintervention is primarily based on statisticalevidence concerning the incidence ofaccidents and fires in particular dwellingtypes, instead of on the causes of thoseevents and how they and their consequencescan be limited.4

2 Full costs have to be calculated. Importantly,they should include spillover effects. Forexample, raising the costs of providing onetype of housing may make it less profitable toprovide, so that supply falls and rents rise.Some current or potential tenants, therefore,may have to choose worse and more riskyhousing options.

3 An important cost consideration is that thegeneral impact of regulatory uncertainty maylower investment throughout the privatelyrented sector. Introducing regulations in onepart of the privately rented sector may worryinvestors in other segments that governmentscannot be trusted to let free markets operate.The risk of regulatory threat may then deterinvestors, thereby restricting supply andraising rents throughout the market-basedprivately rented sector.

4 It cannot be assumed that regulatoryintervention will necessarily work, so anypotential benefits may need to be heavilydiscounted to derive a realistic benefitassessment. With regard to HMOs, forexample, the Housing (Fire Safety in Housesin Multiple Occupation) Order of 1997 hadalready placed an additional duty on localauthorities to ensure that fire safety measureswere adequate in HMOs of three storeys orabove. Yet, the current Housing Billlegislation has by implication deemed thatearlier measure to be ineffective. What is theevidence that the proposed measures will doany better? With limited monitoringresources, for instance, it may be difficult to

46

ensure licensed premises are run in theappropriate manner.

Moreover, raising the hurdle for being alegitimate HMO landlord may encourage theunscrupulous to be illegal ones. To say that theywould be punished by fines when caught andcould not pursue tenants through the courts fornon-payment of rent may be weak deterrents.Criminals (which such landlords would be)generally have other means of encouragingpayment and stopping informants. This isanother risk of regulation: putting thingsbeyond the law can encourage the existence ofsome appalling housing situations.

Dwelling condition and repairs:engineering versus economicapproachesAccording to the English House ConditionSurvey 2001, the privately rented sectordisproportionately has more cases of unfitnessand disrepair than the housing stock as a whole.Ten per cent was classified as unfit in 2001, halfdid not meet the decent homes standard andmore than a fifth required more than £10,000worth of repairs – with 14% in need of £5,000or more urgent repairs.

Though there has been some improvement inthe average numbers in recent years with theletting of better quality dwellings as a result ofthe growth of the market-rented sector, thesedata suggest that there are significant problemswith repairs in the privately rented sector.Should this be a matter of general concern withthe operation of the privately rented sector?

There are several issues of interest here:

1 Concentration of disrepair in certain parts ofthe privately rented sector

The privately rented sector was argued inChapter 2 to consist of a series of segmentedmarkets, so it may be inappropriate to averagedisrepair across all parts of it. Disrepair is likelyto be disproportionately concentrated in

particular market segments, particularly thecheaper and low demand area parts.

2 Is there an optimal level of repair?The measurements in the English HouseCondition Survey are based on a pre-defined setof physical standards. Fully repaired is when therepair criteria are all met. Fitness exists whenthe fitness ones are passed. The same occurswith decent standards and so on. Thesemeasures are devised by well-meaningcommittees, but many lay-people would findcertain elements of them arbitrary andunnecessary. The standards are also allembracing, whereas some dwellings may be fitfor some people’s housing circumstances butnot others. The generalist approach will befurthered with the introduction of the HousingHealth and Safety Rating System5 in the 2004Housing Bill, which will replace the unfitnessand repair standards with a broader set ofmeasurements identifying problems with qualityaspects of the privately rented sector and willbe given a prominent place in the next EnglishHouse Condition Survey.

On the basis of such pre-defined standards andusing trained professional eyes, detailedhousing surveys are undertaken. The finalaggregate numbers on standards and repairlevels are then estimated by extrapolating fromthose sample surveys. Here, this will be calledthe engineering approach.

An economic definition of fully repaired wouldtake a somewhat different tack. A building isoptimally repaired when the marginal benefits ofdoing so equal the marginal costs of repairingand maintaining a structure to a particularstandard. That standard may well be lower thanthe engineering standard, because keeping ahouse fully repaired in engineering terms maysimply not be worth it. Disrepair in economicterms, therefore, may not actually be a socialbad but actually a social good, because it showsthat resources are not being unnecessarilywasted on excessively high building standardsand, so, can be put to better social uses.

47

3 Private renting or older properties?Chapter 2 highlighted the fact that privatelyrented dwellings are often in old properties, farmore so than in other tenures, and that theseproperties tend to be inner city ones, ratherthan ‘chocolate box’ ancient gems lovinglyrestored and preserved by their owners. Thisprofile may help to explain part of the higherrepair needs in the privately rented sector.

4 Market outcomes lead to variations inbuilding quality

Rational responses to housing affordabilityproblems may be for tenants to accept lowerquality housing and for landlords to supply it. A lower quality for a lower price is a commonfeature of many markets and housing is noexception. Low price-low quality outcomes maybe particularly appropriate for some tenants,because many of them will move on again withina short period of time. Living in a lower qualitydwelling for a limited period may help somepeople afford a better home later on or make itpossible to move into a locality at a lowtransaction cost and subsequently search for abetter dwelling.

To impose high minimum standards in housing,consequently, is to close off housing supply andhousing options for people. Housing standardsmay be raised in this way, which looks good inhouse condition surveys, but it also leads tohigher rents and less overall housing inexistence. Moreover, if housing standards were amoving target, with the rules ratcheted up everytime an average improvement was recorded or ahousing-related tragedy occurred, this wouldgenerate general investor fears that repair andimprovement bills might be raised even higher.By creating broad uncertainty, the negativeimpact on landlord investment could be great.

At the end of such a process, a small, thoughpristine, privately rented housing stock wouldbe left. The economic and social cost of suchan outcome would be unfortunate andenormous – and everyone would wonder whereall the housing went.

5 The importance of market-based theoryQuality control, in other words, has a similarimpact to rent control, particularly over thelong-term, in that it can lead to a drasticreduction in the amount of rental housingsupplied. The resultant limited supply wouldcreate shortages and raise general rent levels.The paradox, therefore, arises of costs beingborne in higher rents by those tenants whoselandlords already provide high qualityaccommodation. General housing affordabilitywould also worsen.

Such second round implications of trying tocontrol housing quality are rarely brought intodiscussions about the issue. However, sucheffects are vital to an understanding of the fullimpact of quality regulation.

A similar point could be made about improvingsecurity of tenure, which is sometimes proposedfor the privately rented sector. Stricter securityof tenure rules would raise costs for alllandlords, which would lower supply and,thereby, increase rents. All tenants would loseout from the higher rent charges and only a fewwould actually benefit from the extra security –and many of them would be quite well-off andnot in particular need of such a measureanyway. It would be highly unlikely that theoverall beneficiaries would be the low incomeand the homeless. More probable would be thatsuch groups would be some of the worst affected.

Economic theory, thus, has the advantage oflooking at the wider ramifications of housingmarket interventions and the feedback effectsthat they create. Though often regarded as anapproach that is dismissive of social impacts, itcan often identify the true social consequencesof events which much housing policycommentary tends to gloss over or ignore.

It is particularly noticeable with respect tohousing repair issues that the economicapproach has been almost universally ignored.More commonplace is a theory that putsemphasis on a moralistic behavioural distinction

48

between types of landlord, which can bedubbed the theory of the good, the bad and theignorant.

This theory assumes that there are ‘good’landlords that automatically repair up to thebest engineering standards, simply because theyare good. In contrast, ‘bad’ landlordsdeliberately do not repair in order to makemore money. In between are the ‘fallen angels’:the ignorant ones that would be good if theyknew how to, but know nothing about repairs.

This faulty theory amazingly is promoted inofficial government documents. For example, itunderlies the commentary in the English HouseCondition Survey 2001 Landlords Survey. Thatsurvey report includes, for example, thefollowing observation: ‘...more commerciallyoriented owners of poor condition propertieswho could be described as ‘milking’ them forrental income’. Landlords that turn intomilkmaids are clearly to be regarded as bad, yetthat interpretation hardly counts as unbiasedfactual reporting of house conditions.

An economic analysis brings out the weaknessof such interpretations of the causes of housingquality. Of limited relevance in it are matterslike landlords’ knowledge of building science;whether they have found an honest, localbuilder; or whether landlords are good or badpeople. Rather, the conclusion would be thathousing standards lie on a continuum that isdetermined in particular locations by the cost ofproviding buildings of specific quality levelsthere and tenants’ relative willingness to payfor them.

Three views of housing repairs: whichis right?The levels of disrepair reported by buildingsurveyors are much higher than the views ofeither landlords and tenants. This may be dueto physical factors – not all building problemsmay affect the actual living accommodation oftenants, for instance. Or it may be due to

ignorance, a landlord unskilled in buildingmatters might mistake a serious repair problemfor a minor one.

All the same, some of the reported differencesbetween the three views may be identifying adifference between engineering standards andcost-benefit repair optimisation processes. Theviews of landlords and tenants over repairscertainly should not be discounted. They afterall are the ones that strike bargains over priceand quality in the privately rented sector.

When interpreting survey results, because of thenature of the landlord-tenant relationship it isto be expected that landlords are likely to giveoverly positive views about events associatedwith themselves and that tenants are likely tohave overly-negative views about landlordbehaviour. So, perhaps, the truth may liesomewhere in the middle.

Combining the results of the Survey of EnglishHousing and the English House ConditionSurvey, the following views on housing repairscan be found. Only 3% of landlords and agentssay there are serious repair problems. Incontrast, 22% private tenants say repairs areneeded, but only 10% of them are actuallydissatisfied with repairs. Moreover, only 5% ofprivate tenants are very dissatisfied with theirlandlord overall.

These data suggest a ball park figure for theconsensus view of repair problems in theprivately rented sector on the part of landlordsand tenants. The difference between tenants’views about repairs being needed and their levelof dissatisfaction with the repairs service may bedue to a pipeline effect. That is, things have notbeen fixed yet, but they expect it to happensoon. In the same vein, if repairs were a majorissue, it would be expected that tenants wouldbe very dissatisfied with their landlord, thoughonly 5% say they actually are. The differences inlandlord and tenant views, therefore, are not thatgreat – especially in relation to the much higherfigures reported in house condition surveys.

49

The juxtaposition of these results, therefore,highlights a massive difference between landlordand tenant views, on the one hand, and housecondition survey results, on the other hand. Thissuggests that repair problems in the privatelyrented sector are not as great as they may seemto be. It certainly does not indicate the need forwidespread regulation to raise housing qualitystandards throughout the tenure.

This is not to say that issues like landlordignorance and the difficulty of finding decentrepair and maintenance builders areunimportant. Rather it is to put such factorsinto perspective. Some improvement in housingquality standards may result from betterlandlord knowledge or easier means of dealingwith good quality building services, but theprivately rented sector is not uniquely afflictedby such features. They do not, for instance,differentiate the privately rented sector fromsimilar problems in owner occupation and, so,do not represent convincing candidates toexplain why measured repair levels seem to beso much worse in the privately rented sector.

Narrowing the costs of good andlower quality housing throughtax reformIt was argued above that quality differencesprimarily arise because of differences in thecost of providing housing of distinct qualitylevels and variations in households’ willingnessto pay for them. On the supply side, thedifference in the supply cost between better andworse quality is determined, first, by the relativebuilding costs associated with different states ofrepair and, second, by the scale and costs ofperiodic improvements and modernisations todwelling facilities.

In the UK, both housing repair andimprovement costs are subject to VAT taxation.Housing rents are not subject to VAT, so thatmany landlords are not registered for VAT, asthey have little or no VAT liability. This meansthat most cannot claim back the VAT imposed

on repairs. It is also impossible to reclaim VATon improvements in all cases, because they arenot deemed to be necessary business costs. Inaddition, expenditure on building improvementworks cannot be depreciated in the UK andoffset against tax liabilities. These tax effects,consequently, create an undesirably large wedgebetween the cost of maintaining housing atlower and higher quality levels.

The UK tax system with respect to the privatelyrented sector, therefore, must bear someresponsibility for lower than desired housingquality standards. It seems particularly ironicthat one part of government is exacerbatinghousing quality problems through the taxsystem, while another is trying to improve themthrough extensive and expensive measurementand regulation schemes and large expenditureson urban renewal programmes. The case forlower taxation on repairs and improvementsseems strong.

Some countries allow substantial depreciationallowances for landlords of rental property:notably, Germany, France, Australia and NewZealand. Though not necessarily ideal taxbreaks, these general depreciation allowancesdo have a positive impact on housing qualitystandards in the privately rented sector bygiving landlords an incentive to keep theirproperty at a high standard. The size of thedepreciation allowance is determined by thevalue of the property owned. As housing qualityis an influence on the value of a dwelling, thesedepreciation allowances therefore encouragelandlords to keep up the quality of theirproperties. The depreciation allowance, in otherwords, lowers considerably the net costdifference between owning a good and lowerquality dwelling in favour of the betterquality one.

The supply route to improved rentalhousing qualityAnother way to improve privately rentedhousing quality is to increase the overall supply

50

of housing, because that creates greater choicein a housing market. Landlords would then haveto compete more fiercely to attract tenants andthose with poorer quality dwellings would find itless easy to attract tenants.

Extra housing supply not only reduces thegeneral price of housing, but also lowers theprice differential between houses of differentquality. This makes it cheaper for landlords tooffer a better product as the cost differentialbetween a good and lower quality dwellingnarrows.

In summary, increasing housing supply inducesgreater competition and lowers pricedifferentials. It, therefore, is one of the bestremedies for poor quality housing as it drivesout the worse and raises standards. Regulationof quality does the opposite. It reduces housingsupply and thereby lowers landlord competitionand raises the cost differential betweenproviding good and poor quality housing.Regulation paradoxically can actually worsenoverall housing quality rather than improve it.6

Two exceptionsThe arguments above have assumed thatmarkets work, which is mainly the case in theBritish housing system when they are allowedto. Yet, there are two situations where marketsdo not seem to work well: areas of low housingdemand and privately rented housing that isassociated with housing benefit.

Low demandIn areas of low housing demand, common inmany northern cities, housing supply isseemingly plentiful but, in fact, it is not.Neighbourhood and dwelling quality problemsmake the overall housing supply package shrinkto insignificance. Low demand, therefore, isassociated with low, quality adjusted, supply. Inmarket terms, the housing is worthless – thoughthe land on which it stands is probably not – sodemolition is usually the best answer. Policymakers and commentators, however, often baulk

at what they see as the waste of demolishing‘good’ housing. In fact, the waste more likely isin spending money to retain what is not wanted.

Areas of low demand, in addition, probablydistort the aggregate privately rented sectorrepair statistics, which include such housing.The housing in such localities may be classifiedas private rental property, even when no-onewants it, and will generally be in a poorcondition.

Housing benefitHousing benefit in the privately rented sectormay lock tenants into particular housing,particularly when the rent is paid directly to thelandlord. The bargaining position of the tenantas a result may be weak and the landlord mayconsequently lower the quality of the housingprovided below that normal for the rent level, asthey know the tenant cannot move easily.

This situation may well arise in practice. But, aswas argued in Chapter 3, the housing benefitpart of the privately rented sector is not a truemarket but one controlled by housing benefitrules. As such the criticism is of those rulesrather than the rental market itself. Such weaktenants’ positions certainly suggest that themovement towards housing vouchers mightimprove housing benefit tenants’ negotiatingpower. Furthermore, there seems to be an onuson benefit providers to make periodic randomvisits to their clients’ dwellings to ensure thatcontractual conditions are being honoured.

51

1 They include the Housing and Landlord and Tenant Acts1985 and the Defective Premises Act 1972.

2 Housing policy textbooks, even as late as the mid-1990s, were dismissive or pessimistic about the likelysuccess of the 1989 deregulation measures. See, forexample, P. Balchin, Housing Policy 1995 and J. Hillsed. The State of Welfare 1994.

3 Home and Leisure Accident Report. Summary of1998 data, Department of Trade and Industry, London,2000.

4 The housing provision aspects of fire accidents, forinstance, have been linked to a low proportion offatalities in HMOs, see N. Cowell, An Analysis of FireFatalities in Houses of Multiple Occupation 1996-2003, National Landlords Association, 2003.

5 See ODPM website, www.odpm.gov.uk, and “TheHousing Health and Safety Rating System – a newmethod of assessing housing standards reviewed”,J. Stewart, Journal of Environmental HealthResearch, Volume 2, Issue 1, July 2003.

6 A corollary of this argument is that the UK has a poorerquality housing stock than some of its Europeanneighbours because housing supply is more limited herethan in those countries.

52

IntroductionThis chapter examines the interrelated issues ofresidential investor returns and taxation. Iflandlords cannot generate long run risk-adjusted returns that are at least as good asthose that can be earned by investing in otherassets, they will not invest in the privatelyrented sector. Similarly, if tax breaks andsubsidies are more attractive in other tenures,potential tenants will choose to live in them – ifthey can. Returns and taxation are consequentlycentral to the future of privately rented housing.

This chapter points out that landlords face anumber of unique running costs that do notexist in owner occupation. As a result, rentshave to be higher than equivalent owneroccupation costs in order that landlords canearn a competitive return. This is one of themain reasons why homeownership is a moreattractive tenure for many households. Thissuggests that the privately rented sector cannotbe expected to increase significantly at theexpense of owner occupation, even when taxregimes are neutral in their impact on onetenure or another.

Traditionally, the tax breaks offered to owneroccupation in the UK were much more attractivethan those available in the privately rentedsector, but over the last decade that haschanged significantly. The current tax positionof these two tenures is investigated below, aspart of a general discussion of taxation and theprivately rented sector. Some reforms to the taxregime in the privately rented sector aresuggested.

There is not a detailed discussion of therelationship between taxes and subsidies inprivate renting and social housing, because theissues have been considered in earlier chapters.

In general, however, the subsidies and taxbreaks existing in the social rented sector meanthat most low income tenants only live in theprivately rented sector if they cannot get decentsocial housing in the neighbourhood where theywant to live.

Higher running costs in the privatelyrented sector Landlords face a variety of annual costs, apartfrom those associated with servicing loans andthe opportunity cost of their own capital.

One of the greatest influences on net rentalincome after running costs is the prospect ofhaving no tenant (a void) for significant periodsof time.1 Average voids have recently beenrunning at about a month a year for dwellings inmarket renting.2 This average vacancy period ispartly determined by the current balancebetween demand and supply, but to a greatextent reflects the mobility of tenants and theassociated churn of tenancies. Landlords cannotinstantly match their dwellings with a newtenant when a previous one leaves and,therefore, have to accept times when theirproperties are empty. Mobility in the privatelyrented sector has a hidden cost, in other words,in the form of vacant periods when no-one livesin such dwellings, which seems currently in thefree market sector to be around 8% of the timea rental dwelling is in existence.

Non-payment of rent is another factor, either inthe form of arrears or bad debts from tenantsthat have left without paying all of the rent. Inaddition, there are costs associated with agencyfees, or other charges for finding new tenants,plus on-going property management fees andother costs such as keeping accounts. The wear-and-tear on soft furnishings, equipment and

CHAPTER SIX

Returns and taxation

53

internal decoration has also to be taken intoaccount. Table 6.1 shows a variety of othercosts. Altogether these costs typically accountfor 30 to 60% of gross potential rents.3

Such high running costs help to explain whyprivate landlords are keen to keep good tenants.A long let to people that minimise wear-and-tear costs and promptly pay their rent leads tofar lower costs.

Table 6.1 Non-capital residential letting costsVoids*Bad debts*Net arrears*Letting fees*Property management fees*Service charges†Ground rent†Building insuranceContents insuranceMaintenanceRepairs (minor)**Replacement/repair soft furnishings and contents**

*renting only †flats only **higher in renting

These running costs also show why investors inresidential property need far higher grossreturns on rents than do investors incommercial property. Commercial property doesnot lead to such high management costs,because the risks associated with rent arrears,bad behaviour and high tenant turnover are farless. Commercial leasing contracts are alsostructured to induce particular forms of tenantbehaviour, such as limited turnover. They leadto an apportionment of risks away from landlordto the tenant in contrast to the residentialsector – as occurs with commercial long leasesand upward only rent reviews. Such contractsare generally unfeasible in residential letting.Commercial landlords can therefore operate atfar lower gross margins than residential ones.

Landlords’ running costs also help todemonstrate why from a household’s point ofview renting typically is significantly moreexpensive than home ownership. Homeowners

are both their own landlord and tenant, bydefinition, and so face no voids nor arrears.Their property management costs are noticeablyless, because they do not have to deal with hightenant turnover. Finally, wear-and-tear costs arelikely to be lower for owner occupiers, becausethey are reduced by the incentives of lookingafter your own property. The extra costs borneby landlords consequently have to be reflectedin the rents they charge for an equivalentproperty to an owner occupied one in additionto the similar pre-tax capital costs of housepurchase, structural maintenance and repair ineither tenure.

Such extra running costs for landlords aresubstantial, as already noted, and, in the main,constitute the economic deadweight cost ofprivate renting. They are deadweight in thatthose costs would not exist if housing werelived in a different form (i.e. as owneroccupation).4

One way of looking at these cost differences in theaggregate is to recognise that owner occupationconfers significant economies over renting as aform of managing the existing housing stock andpeoples’ housing circumstances. Those costdifferences are generally not included in housinguser cost of capital comparisons between rentingand owning, but help to explain the growth ofowner occupation.

Net rental returnsHaving taken account of their running costs,landlords’ returns consist of net rental incomeplus any capital gains minus taxes. There ismuch talk about whether current net returns inthe privately rented sector are sufficient tosustain it over the long-term. A significantcontributor to residential returns has beencapital growth in recent years, with house pricesrising well above trend levels. Such capital gainreturns, of course, cannot continue forever. So,it could be argued that landlords are likely toface lower returns in the future, leading to manyquitting the sector.

54

This may not be such an important issue for thecurrent size of the privately rented sector as issupposed, because total returns currently areprobably well above their long-run equilibriumlevel, particularly for geared investments. Theimpact of gearing on residential investor returnscan be seen in the ARLA returns data for theend of 2003.5 The total return for a cashresidential investor was 11-12%, but for one thatborrowed 75% of the initial capital coststhrough a mortgage, the return was double that.These returns are very high for what is normallyclassified as a relatively low risk, safeinvestment class. A decline in returns from suchlevels should not lead to massive retrenchment,especially if more residential investors takeadvantage of gearing possibilities.

Taxation and its consequences

Tax distortionsMost taxes are non-neutral in the sense thatgovernments by imposing taxes distort themarkets from which they collect their revenue.Subsidies can have similar effects by alteringrelative production costs or the prices ofvarious consumption possibilities. Taxation andsubsidy have long been known to distort activityto a particularly large extent in the housingmarket and, because housing is such a largeelement in personal expenditure, suchdistortions have widespread economic andsocial impacts.

Government fiscal policy distorts housingprovision in two ways. First, by encouragingexpenditure on housing in relation to otheractivities. Second, by altering the relative costsof living in particular tenures. Private rentinghas been on the bottom rung of the housingpolitical ladder for many years and,unsurprisingly, the tax and subsidy system haslong tended to be biased against it.

Yet, before examining taxation problems in theprivately rented sector, it is important toremember that housing as a whole has a light

tax load and a heavy subsidy lift compared tomany other activities. So, any problems withprivate renting as a tenure need to be temperedby the fact that rental housing consumers (and,hence, residential investors) gain from thegeneral housing beneficence of governments aswell as those living and investing in othertenures.

Justification for treating housing so favourablyis often couched in the moral terms of shelteras a basic human need but, as there are few cut-off points, most of the tax benefits apply rightup the scale to profound luxury and locations ofthe highest status and convenience. Installing asplendid conservatory or moving to a smarterlocation for an owner occupier, for example, ismore tax efficient than buying a large car orexpensive TV – though why is hard to justify.The distributional implications of many taxbreaks are also regressive in that the better-offand the more expensive homes often gain themost.

The light tax load has been compounded byquirks in the tax system. An important factor inhousing is that second-hand durable goods arenot covered by VAT. Consequently, the housingservices released from the existing stock ofhousing and the benefits of living at somelocations rather than others incur no VATliability, whereas many other consumptionactivities do. Repairs and improvements arecaught within the VAT tax net, however, soolder dwellings eventually end up generatingmuch VAT. The outcomes are perverse. Up-graded houses in the grandest locations rise invalue as a result of repair and improvements byfar more than do those in the worst ones. Evenso, the VAT tax liability per £ spent is thesame.6

Unlike most other EU countries, new housing isnot subject to VAT. As many other consumptionactivities incur 17.5% VAT, the incentive tosplash out on new housing rather than otherconsumption activities in the UK is high.

55

Possible tax reforms The lack of consistency in the housing taxsystem and potential feedback effects, as well asthe generosity of the overall tax burden, createthe need for complex strategies for reform thatinvolve ‘second-best’ responses. Imposing VATon new build, for example, might discouragehousing supply. Yet, if that argument is right, acorollary should be that VAT is not imposed onrepairs and improvements of older dwellingseither, in order to avoid distortions in favour ofnew build over older buildings andconservation. This would be of particularbenefit in the privately rented sector whereolder housing and lower repair levels prevail(see Chapter 5). However, the recent BarkerReport7 argued against such a policy suggestingthat there would be a large deadweight loss offoregone taxation on works that are alreadycarried out and that better-off homeownerswould benefit the most from the removal of VATon repairs and improvements.

A useful principle when contemplating loweringthe tax generosity of housing should be that it isbetter to tax housing consumption than housingsupply – though development land is anothermatter. Housing supply is widely accepted asbeing far too low, for a variety of reasonsexplored in the aforementioned Barker Report.Taxing supply makes the matter worse. For thisreason, a property tax based on annual propertyvalues is one of the better tax reformssuggested in recent years.8

A stamp duty anomalyThe taxation of private landlords in severalrespects is similar to that of businesses ingeneral in that costs can be offset againstrevenues before tax is charged, but differs inseveral ways. In particular, residential propertycompanies face one expensive tax quirk withrespect to other companies, because of theirline of business and the nature of stamp duty onresidential property.

The problem arises with stamp duty thresholds.The value of individual rental dwellings may

well fall within one of the lower tax bands. Yet,if purchased jointly when a block of flats isbought, a higher threshold value is triggered onthe whole value of the acquisition. Taxationpolicy, thus, discriminates against blockpurchases and is at variance with governmentpolicy which aims to increase the role of largerproviders within the privately rented sector.This measure particularly discourages largerinvestors (and any potential future PIF), becausethey are only likely to be interested in biggersingle investments, such as the purchase ofresidential blocks of flats.

Private individual landlords and mortgageinterest tax reliefPrivate individual landlords pay tax on theincome, net of costs, from their residentialinvestments at their marginal tax rates. Thebiggest beneficiaries from such tax reliefs arethose on higher marginal tax rates. The better-off consequently can gain more from beinglandlords than those on lower tax levels,particularly if they borrow in order to recoupinterest tax relief.

Since the recent abolition of mortgage interesttax relief for owner occupation, investing inprivately rented housing has become a relativelybetter way of maximising tax benefits, especiallyfor higher marginal rate individuals. A UKtaxpayer, for example, cannot claim tax relief byborrowing in order to invest in equities but theycan by taking out a buy-to-let loan. This tax gainmeans that the privately rented sector inequilibrium can operate on a lower pre-tax returnthan other less-tax efficient investment vehicles.

In reality, as early chapters showed, the privatelyrented sector currently is under-mortgaged withmany landlords not taking advantage of taxreliefs. This is likely to change in the future asmore residential investors realise the gearingand tax benefits of relatively high levels ofborrowing. An implication, of course, of thisforecast increase in mortgage borrowing is thatthe Exchequer’s tax take from the privatelyrented sector should fall significantly over time.

56

Individual landlords and running costsIndividual landlords that use agencies tomanage their properties have to pay VAT onthose services but can offset the costs againstrental income and, thereby, reduce their taxliabilities. Landlords that do their ownmanagement are not liable to pay VAT on thoseservices, but neither can they offset their effortsagainst tax. There have been objections to thissituation as it seems to imply a tax on personaleffort.9 However, the picture has another side. Ifmanagement DIY landlords could includenotional management costs as tax deductions,symmetry would suggest that they are thengenerating an income-in-kind for themselves. Ifthis income-in-kind were taxed that would offsetthe benefit of the proposed tax deductibility ofDIY management costs. Both landlords and theInland Revenue would be involved in greateradministrative effort necessary to monitor, claimand process the in-kind cost claims and incomecharges, with little net effect on overall taxation.The deadweight costs of such measures,therefore, suggest that they should not beimplemented.

Individual landlords and capital gains taxationPrivate individual landlords are subject tocapital gains taxation (CGT), less permittedindexation and costs,10 excluding initialallowances,11 when they sell properties. CGT ischarged at the person’s marginal tax rate. Thereis also a taper on CGT charges: after 2 years ofownership, 5% of the tax is waived for eachsubsequent year of holding up to a maximumdiscount of 40% in the tenth year. This taperobviously encourages landlords to hold ontotheir properties until they maximise the taxrelief.

Taxable capital gains arise from any increase inthe value of the property due to general houseprice inflation, any local relative increases inhouse values and, also, the impact on the valueof any property of repairs and majorimprovements to it. The indirect imposition ofCGT on repairs and improvements is anothertax disincentive on private landlords to repair

their dwellings. It drives a further wedgebetween the costs and benefits to landlords ofproviding high and low quality property.

In contrast to landlords, owner occupiers arenot subject to CGT on sale after a minimum sixmonth period of living in their principalresidence. This tax break currently costs theTreasury £10.5bn a year – though this figuremay be exceptionally high because of thecurrent house price boom in the UK.12 So,homeowners do not face the same CGTdisincentive to repair and improve that exists inthe privately rented sector.

There is another anomaly of CGT in theprivately rented sector. Unlike companies,private landlords cannot rollover this liabilitybetween assets, because for them residentialproperty is not counted by the Inland Revenueas a ‘business asset’.13 This is an importantdisadvantage as it discourages personallandlords from trying to optimise theirresidential portfolios or from changing them inresponse to shifting patterns of demand.Removing the anomaly is complex, however,given the far higher taper relief currently givenfor business assets.

The tax position of landlords and owneroccupiers comparedThere are several differences between the taxsituation of private individual landlords andowner occupiers.

Two differences are mirror images of eachother: landlords do not get relief from CGT andowner occupiers cannot offset interest paymentsrelated to their properties against tax liabilities.Whether one or the other is a better benefitdepends principally on the relative magnitudesover time of interest rates and annual increasesin house prices. Ignoring allowable deductionsand assuming that general price inflation affectsinterest and house price inflation rates thesame, the benefits of either depend on therelative rate of change of interest rates andhouse price inflation. The trend rise in real

57

house prices is high in the UK at between 2.5%and 2.7%,14 so the CGT relief benefits those UKowner occupiers more than those in lower trendhouse price inflation countries where such taxbreaks are also given to owner occupiers. Yet,over the long-run, the benefit of each type oftax break is anyone’s guess; though the fact thatCGT is imposed at the time of sale, with aninitial tax free element and contains a markedtime taper, whereas interest tax relief (ITR)applies to interest rates in full and is recoupablein the current tax year suggests that over thelong-run ITR may be more advantageous.

The short-run situation, nevertheless, can bedifferent from the long-run one. Both houseprice rises and interest rates go through cyclesand those cycles tend to move in oppositedirections. Since the mid-1990s in the UK, forexample, nominal interest rates have beenfalling significantly, while house prices havebeen rising rapidly. The pendulum consequentlyhas favoured CGT as the better tax break overthe past decade and, hence, benefited owneroccupation over private renting. Prospects forthe future suggest that the pendulum may swingthe other way, with higher nominal interest ratesin prospect and lower annual house price risesfor several years to come. This means that therelative tax benefit could now be swinging backin favour of privately renting over owneroccupation.

The other main tax difference between theprivately rented sector and owner occupation isthe tax treatment of rental income. On the onehand, landlords have to pay ‘Schedule A’ tax ontheir rental income net of current costs.15 Owneroccupiers, on the other hand, do not pay such atax on the (imputed) rent they as ‘tenants’implicitly transfer to themselves as‘owners/landlords’. The tax difference perequivalent dwelling consequently is currentlysubstantial, especially as many landlords do notoffset this tax by gearing in the ways describedabove. However, as borrowings become morecommon in the privately rented sector, thedifferential between the tax efficiency of the two

tenures will narrow, because significantlysmaller shares of landlord’s gross rental incomewould be subject to tax.

The tenure tax burden, in summary, is a jointproduct of tax rules and how decision makers ineach housing tenure respond to them, but theabsence of Schedule A tax favourshomeownership over renting. If landlordsbecome more financially and tax savvy, therelative tax incidence of the two tenures shouldbecome narrower, especially as over the long-run the benefits of ITR are likely to outweighthose of non-imposition of CGT. Thisconclusion, however, does not hold over thehousing market cycle, when tax effects are likelyto cause shifts in both demand and supply ineach tenure – with the direction of those shiftsdepending on relative changes in house pricesand mortgage interest rates.

The potential for tax arbitrageThere are many ways to look at the impact of taxon housing markets. One way is to take astructural view and see whether households asgroups can choose housing investment andconsumption options that minimise their jointtax and housing cost payments. This issue isgenerally known as tax arbitrage.

The simplest way to conceive of this propositionis to imagine two groups of households, onepaying a higher marginal rate of income in taxthan the other and, then, think about the waythe higher tax group can use tax breaks inprivately rented housing to make housingavailable to lower taxed groups. The latterindirectly share in the tax benefits because thetax-induced extra housing supply leads to lowerrents for them.

In the current UK tax situation as describedabove such tax arbitrage is possible, particularlywhen those on higher tax bands borrowmortgages to fund the purchase of dwellingsthat they rent out to those on lower tax bands.In this way, the total tax take of both groups isminimised and the higher band households

58

share some of their tax breaks with lowerincome households through reduced rents. Thelower income groups, consequently, have lessincentive to move into owner occupation.16

The prediction of the tax arbitrage model is thatbetter-off households will provide rentalhousing for lower income groups, who may havelower incomes simply because they are younger.Unfortunately, the empirical evidence to datehas not offered much support for the taxarbitrage model. First, as noted above, therecent phases of the housing market andinterest rate cycles have strongly favoured theCGT break offered to homeownership. Second,financial liberalisation with respect to theprivately rented sector is still in its infancy, sothe borrowings necessary to support the taxarbitrage effects have not been sufficient tohave an impact on privately rented as a whole,at least in the UK. Third, the running costs ofprivately rented accommodation are significantlygreater than those of owner occupation and thistends to counteract potential tax arbitragebenefits. Management problems, moreover, areoften greatest in the cheaper parts of theprivately rented sector and this is the sectorwhere tax arbitrage effects are likely to be themost pronounced. Tax arbitrage in the privatelyrented sector consequently seems to be more ofa theoretical possibility than a practicaloutcome, although it may become moreimportant as house price growth moderates inthe UK and mortgage interest rates rise.

1 Turnover rates of tenants were examined in Chapter 3.

2 ARLA Review and Index of Returns on ResidentialInvestment.

3 These figures are derived from a variety of sourcesincluding ARLA, London Residential Research and A.Crook et al Investment Returns in the PrivateRented Housing Sector¸ 2002, British PropertyFederation, London.

4 The deadweight cost calculation can only be imprecisebecause the high turnover rates in private renting to adegree provide economic benefits.

5 As actual landlord returns cannot be easily measured,there are several published notional return calculations.The actual level of returns recorded in them, of course,

depends on the assumptions made. Those released byLondon Residential Research, for example, typicallyproduce more pessimistic returns data than those shownabove.

6 Though a vibrant no VAT residential improvement andrepair industry has grown up, either because builderproprietor turnovers are below the VAT threshold orbecause they fail to charge VAT.

7 K. Barker Review of Housing Supply – Final Report,HM Treasury, 2004.

8 See J. Muellbauer Safety in Property Tax,http://www.housingoutlook.co.uk/Papers/ft0702.html.

9 See, for example, the recent JRF/Shelter report FairDeal for the Privately Rented Sector, 2003.

10 Indexation is allowable for the effects of inflation up toApril, 1998; permitted costs are items such as the legaland agency costs of acquiring the property.

11 The first £8,200 of the chargeable gain is not subject totaxation in 2004-5. This allowance is indexed andchanges annually.

12 Treasury Budget Report, 2004. A tax take and taxrelief balance sheet is not published for the privatelyrented sector, so the magnitudes of tax for it areunknown.

13 See Capital Gains Tax, Inland Revenue PersonalTaxpayer Series CGT1.

14 See Barker Report, ibid.

15 ‘Rent a room’ landlords, however, have a £4,250exemption on annual gross rent. See Letting and YourHome, Inland Revenue Personal Taxpayer Series IR87.

16 Technical demonstration of this tax arbitrage effect anda discussion of actual outcomes can be found in G.Wood ‘Are there tax arbitrage opportunities in privaterental housing markets? Journal of HousingEconomics, 2001, 10, 1-20.

59

IntroductionThe previous chapters have undertaken alengthy analysis of the privately rented sector.This chapter summarises the main points andthe conclusions drawn from them. Policyrecommendations are identified. A final sectiondraws the major policy themes together.

Summary and conclusionsThe main points of this report are summarisedbelow. Where they have a direct policyconsequence an asterisk is assigned to them.

Overview• The overall size of the rental sector in the UK

is around the international average, butexceptional reliance on social housing makesthe privately rented sector one of the smallestat 10-12% of the housing stock.

• The revival of the privately rented sector overthe past fifteen years has been underpinnedby deregulation that has for the first time inthree-quarters of a century allowed marketforces to work effectively.

• Deregulation enabled the unleashing ofseveral key drivers in the growth of theprivately rented sector: financial liberalisationand competition, rising investor wealth andimproved perceptions of landlordism, risingproperty values, reduced tax breaks for owneroccupation and higher demand caused byrising standards of living, more students andincreasing barriers to entering owneroccupation.

• Privately renting is greatest in South EastEngland, university and seaside towns, andsome large cities – especially London.

• Housing standards are generally good. Few dwellings are overcrowded; a third areflats, but most are houses; the majority have access to gardens or terraces. The privately rented stock is generally older than in other tenures with 60% builtbefore 1945.

• The privately rented sector consists of many segmented markets serving differenttypes of client, such as lodgings, short lets, furnished and unfurnishedaccommodation, unsubsidised and housing-benefit tenants, plus markedgeographic specialisation.

• An expansion of private renting at theexpense of owner occupation has no cleareconomic or social justification.*

• This study has not considered social housingin detail, but it would benefit from operatingmore on market principles like the privatelyrented sector.*

Who rents, with what effects?• Tenants are highly mobile: a third move within

a year, two-thirds within three years.

• Moves are mainly within the privately rentedsector. Cross-tenure moves are notparticularly different from other tenures.

• Private renting now is most commonly atenure of transition in that the majority oftenants will become owner-occupiers at alater stage of their housing careers.

• Most tenants are young adults: almost two-thirds are aged less than 35 and a quarterunder 25.

CHAPTER SEVEN

Summary of findings and policyimplications

60

• The youngest households pay the highestrents and the oldest the lowest.

• Most privately rented tenants are in work insharp contrast to the socially rented sector.

• In terms of household types: two-fifths arecouples, a third singles; sharers represent afifth and single parents a tenth.

• The income distribution of privately rentedtenants is similar to the national distributionof income of all the three main housingtenures.

Future demand• The young adult age cohort is expected to

decline in absolute numbers between 2002and 2016. At the national level, the fall is notgreat but in South East England and London,especially, the fall will be significant at over10%. If this forecast is correct, it could lead toa shrinkage of the privately rented sector.

• Counteracting such population changes islikely to be an increase in the wealth barriersto owner occupation, forcing more youngeradult households to remain in the privatelyrented sector for a longer period. In the shortrun, however, the demand for privately rentingmight temporally fall once the current houseprice boom is over, as some tenants will thenmove into homeownership as house price toincome ratios decline.

• A reform of social housing would increasedemand for private renting.*

• No large-scale increase in the privately rentedsector can be expected in the near-future inthe absence of reform to social housing.*

Mobility and the labour market• Mobility may be high in the privately rented

sector but that does not mean more privatelyrented dwellings would necessarily increaseoverall labour mobility. As most workingpeople are owner occupiers, the country

actually needs more owner occupieddwellings in order to increase labourmobility.*

• Social problems such as divorce and familybreak up also mainly occur within owneroccupation.*

• Housing supply-side constraints are thebiggest factor constraining labour mobility. Inorder to improve labour mobility, extrahousing is needed, in order of priority, inowner occupation, private renting and socialhousing.*

Housing benefit and privately renting• Far fewer private tenants have their rents

subsidised by housing benefit than twentyyears ago, whereas the numbers are roughlythe same in social housing.

• Housing benefit does not work well.Administrative problems are an importantdifficulty. Housing benefit is unpopular withmost private landlords.*

• De facto rent controls in the privately rentedsector were re-introduced with ‘referencerents’ set in relation to housing benefit. Thishas helped to create quality problems.

• The current Pathfinder experiments withStandard Local Allowances seem to beworking, but they do not resolve thefundamental problems of housing benefit.*

• Housing benefit authorities should take on agreater advocacy role to ensure that tenantshave means to complain when landlords donot stick to contractual agreements. Thisshould be applied in both the private andsocial sectors.*

Ownership and investment• Many lettings are owned by private individual

landlords. Corporate landlords havecontinued to withdraw from the sector. Mostinvestors in the privately rented sector,

61

including both private individual andcorporate ones, do so as a sideline to theirprincipal activity.

• There seem to be low economies of scale inprivately rented investment and management,which helps to explain the successful growthof the small investor. Large investors,however, can still prosper.

• There is no reason to expect that smallinvestors are ‘worse’ landlords on averagethan larger ones.

• Landlord-tenant relationships generally arebetter in the privately rented sector than insocial housing. This is because they haveincentives to get on.

• There is a need for improved publicinformation on issues associated with theprivately rented sector for both tenants andlandlords.*

• Tenant deposit disputes are an area ofconcern, but are likely to be best dealt withby private initiative. The Government’scurrent ‘wait-and-see’ policy seems wise asprivate initiatives are currently emerging.*

• The management of lettings and tenancies hasbeen improved by the growth of lettingsagencies.

• The buy-to-let boom has been spectacular,with 400,000 outstanding mortgages by theend of 2003 from zero at the beginning of1998.

• The privately rented sector, nevertheless, isstill heavily under-borrowed, with only a thirdof properties mortgaged. As a result,mortgage lending should grow and is likely to.

• There is no evidence of an impending crash inthe buy-to-let market.

• Proposals to introduce Property InvestmentFunds (PIFs) as tax efficient investment vehicleswill benefit the privately rented sector.*

• PIFs will not generate dramatic changes toprivate renting because of the tenure’s sheerscale at almost 2.5 million dwellings.

• Yet, PIFs will be particularly beneficial forstimulating the provision of new blocks offlats in cities and, where possible withinplanning guidelines, planned communities.

• PIFs could be a useful policy tool in urbanregeneration.*

• PIFs could be expanded into areastraditionally within the remit of the socialhousing sector to provide a cheaper, moreefficiently operated stock and improvedhousing services.*

Licensing and regulation• The introduction of controls on parts of the

privately rented sector, such as houses inmultiple occupation (HMOs), may be well-intentioned but carries the risk of undesirableside-effects.*

• It is not clear that a full cost-benefit analysishas been undertaken of proposals to regulateHMOs. The full impact of HMO licensingdoes not seem to have been assessed, nor isit clear that the desired outcomes will result.There is a substantial risk that the supply ofHMOs will fall significantly in the face ofhigher landlord costs and uncertainties.*

• Safety issues are important, but a strategy ofimposing zero risk in housing provision isneither feasible, economic nor desirable.*

• There is a danger that poor housingconditions in HMOs will still exist even afterlicensing, especially in a criminalisedunderworld of unlicensed premises.

62

• A tendency to introduce regulations whenevera problem is perceived threatens to create aclimate of uncertainty amongst privatelyrented investors as a whole that could damageconfidence and sharply lower overallinvestment.*

Dwelling conditions and repairs• Government surveys paint a picture of a

privately rented sector where dwellingconditions are notably worse than elsewhere,with, for instance, a fifth of dwellings identifiedin 2001 as needing £10,000 of repairs each.

• Landlord and tenant surveys, in contrast,indicate far lower concerns with conditionand repairs.

• Official Housing Condition Surveys take an‘engineering-assessed full repair’ approach tomeasuring housing conditions in all tenures,rather than consider an optimal state of repairfor the housing stock. In doing so theyexaggerate repair problems.

• This approach creates particular biasesagainst the privately rented sector, where thehousing stock is generally far older than inother tenures because of the nature of themarkets served by it.

• It is important not to force landlords to repairup to engineering standards, otherwise rentalhousing supply is likely to decline drastically.*

• Housing benefit agencies should be obligedto undertake random surveys of properties toensure that landlords are providing what isagreed.*

• The taxation of repairs and improvementscontributes to lower rates of repair andmodernisation and should be removed orreduced.*

Landlord returns1. Costs• Operating expenses in the privately rented

sector are high because costs exist that donot arise to the same extent in other tenures.They occur because of the higher transactioncosts associated with substantial tenantmobility, especially the impact of significantperiods of time when dwellings are emptybetween lets; and higher management andwear-and-tear burdens.

• As many such costs do not exist in owneroccupation, it is generally cheaper on a like-for-like basis. This, as well as consumerpreferences, helps to explain why owneroccupation is now the dominant tenure in theUK and most other countries.

2. Returns• Net returns are currently very high in the

privately rented sector because of recenthouse price rises.

• A decline in returns to more normal levels isunlikely to lead to a substantial fall ininvestment, because of the long-term under-representation of residential rental property inmost households’ and institutions’ investmentportfolios.

Taxation• Generally, housing is taxed lightly in

comparison to other areas of consumption andinvestment. There is no obvious justificationfor this, particularly as there are few caps on thetax breaks, so that the better-off gain the most.*

• Taxation policies vary between individuals andcorporate landlords in several key respects.Net rental income, for example, is taxed atprivate landlords’ personal marginal tax rates.For corporates, it is taxed at corporation taxrates and any distributed profits are taxed atshareholders’ dividend income tax rates.Property Investment Funds (PIFs) will createanother tax regime – with only dividendsbeing taxed.

63

• Many corporate landlords, especially sidelineinvestors, because of their characteristics astrusts or public bodies have no corporationtax liabilities. Others can pool residential taxliabilities with the main parts of theirbusiness. Sideline corporate investors tendnot to adopt a commercial attitude toresidential property, but their tax statuspartially shields them from the financialconsequences.

• Tax pooling for individuals is not allowed inthe UK, though it is in some other countries,such as in Australia. It is probably not a goodidea to introduce tax pooling for privateindividual landlords as it may encouragemarket de-stabilising investment activity.*

• There is a stamp duty anomaly for corporatelandlords in that they have to pay the highestrate of duty when buying blocks of apartmentseven though individual units within them mayfall into lower duty bands.*

• Individual landlords cannot transfer liabilitiesfor capital gains tax across their propertyassets, which limits their flexibility inresponse to changing patterns of demand.*

• Arguments that private landlordsunreasonably cannot claim tax-deductions fortheir own management time ignore the factthat currently the income they save fromdoing so is also not taxed as an in-kindbenefit.

• Comparing the tax positions of privatelandlords and owner occupiers, landlords paycapital gains tax (CGT) – though at adeclining rate the longer they hold theirproperties – but can claim interest tax relief(ITR); whereas the opposite situation existsfor homeowners. Over the long-run, ITR islikely to be more beneficial than the CGTexemption, especially when landlords areappropriately geared.

• Private landlords are currently under-geared(leveraged) on average and so lose out on ITRbenefits.

• Homeowners’ imputed rental income is nottaxed, whereas most landlords pay taxes ontheir net rental incomes. This differencerepresents a taxation bias towards owneroccupation. However, if landlords borrow, ITRcan be offset against rental income which cansignificantly lower the bias.

• There is little likelihood that tax arbitrageworks in the UK, whereby better-offhouseholds facing higher marginal tax ratesborrow and let out rental property tohouseholds in lower marginal tax bands.

A policy framework to improve thefuture of the privately rented sector Many policy issues have already been flagged upin the previous section and there is little needto repeat them again. However, there are anumber of overarching themes related toregulation and policy frameworks, taxation andinvestment that still need to be drawn together.

1. Regulation and policy frameworks

Avoid over-regulation and legislative uncertaintyConcern has been expressed in this report thatthe policy climate is moving away from asupport for free market processes in theprivately rented sector to greater control andregulation. It is important not to stifleinvestment in privately rented housing throughan excessive bureaucratic burden. Clear, strongand consistent commitments to a market-oriented privately rented sector need to bemade by policy makers.

Create transparent legislative frameworksComplex legislation only becomes clear in itsconsequences several years after introduction, asthe regulations are tested in the courts andministerial interpretations are made. The

64

measures contained in the current Housing Billwith respect to the privately rented sector are suchan example of complex legislation. In particularly,it may have far reaching consequences for HMOs.In Scotland, in any case, similar measures arealready law. Nevertheless, it is important to limitany further uncertainty that may be generatedamongst landlords to avoid damage to investorconfidence. With this in mind, it is important toensure consistency across local authorities in theimplementation of their new powers and makesure that a light touch rather than a heavy handbecomes the standard guideline.

The need for a positive perspectiveThe revival of the privately rented sector overthe past fifteen years has been an extraordinarysuccess. The attitude of much policycommentary, however, continues to havenegative undertones.

The worst criticism that can be made of aprivate landlord still remains in many quartersthat ‘they are only in it for the money’. Yet, thatis precisely the point about supply in a market-based system – it is there because people wantto make money – and housing is no different.Being in it for the money does not make alandlord a bad person, nor one who treatstenants badly or cheats them. Such behaviours,when they exist, instead mean that landlordsmust have other less savoury motives than meremoney-making ones, because, as was pointedout in earlier chapters, they are not usuallyprofit-maximising ones.

Another common view is that there is always aproportion of landlords that are up to no goodand that tough regulations on all landlords arenecessary to constrain the cowboys. This view isagain misplaced and not universally applied inother lines of business for good reason – manyrules can easily end up imposing more costs onan unexpectedly wide variety of people thanproviding benefits. Some of the issues over theunintended consequences of regulations withregard to the privately rented sector have beenexplored earlier in this report.

A more positive attitude to the privately rentedsector is much needed, particularly amongstlocal authorities. Devolved governments havealso been far quicker to control the privatelyrented sector than to encourage it, despitedeep-seated housing problems, deficits inhousing investment and significantly lower shares of private rental housing than England.

The final policy attitude worry is that in its newsuccessful state, the privately rented sector maybecome a policy instrument to be set targetsand cajoled into meeting them on the strongassumption that the analysis underlying thepolicies is unfailingly correct. History, instead,has shown that policies towards the privatelyrented sector are usually inappropriate.

2. Taxation reforms

Several taxation reforms have been suggested inthis report. Two are related to the purchase andsales of rental properties. These are removingthe stamp duty anomaly associated with thepurchase of blocks of flats, where the individualflats fall into a lower stamp duty band; andallowing individual private landlords to transferCGT liability when they sell and purchaseproperties.

The major tax changes suggested, however,relate to repairs and improvements. As concernexists over repair levels and housing quality inthe privately rented sector, it seems perverse totax them. Building works to individualproperties also generate externality effects byimproving general neighbourhood quality. Theyalso boost general housing supply.

A tax incentive approach to improving housingstandards, furthermore, is a much betteralternative than a regulation approach that triesto force landlords to raise standards. Theregulation route is likely to fail because itimposes costs and uncertainties, whichconsequently lead to a significant withdrawal of

65

investment. The tax break route, in contrast,lowers costs and so encourages extrainvestment.

The proposals are, either:1. To remove or reduce VAT on landlords’

repairs and improvements. The aim would beto encourage further investment in repairsand upgrading the rental stock. Practically,this may prove difficult to implement,however, because of the need to apply taxrules fairly across all activities, as owner-occupiers, for example, would still be chargedfor them.

Moreover, such a reform would face theobjections outlined in the recent BarkerReport (see Chapter 6).

2. To provide other taxation inducements to theupgrading of the conditions and theamenities of the ‘available to the public’privately rented housing stock. In somelocalities, improvement grants already existand some tax reliefs with respect toinsulation and other environmental mattershave also recently been introduced. Yet, thereis a case for offering depreciation allowanceson repairs and improvements in order toencourage landlords to undertake moreinvestment in their existing dwellings.

Such depreciation allowances should ideally beconditional ones. With regard to repairs, forexample, in order to claim the allowancelandlords could be required to show proof ofaverage annual repairs expenditure at least aslarge as some predetermined sum. Thatpredetermined sum could be set at a levelimplying that a dwelling remains in a good stateof repair and vary in amount for dwellings ineach council tax band. The size of thepredetermined sum, the depreciation chargesallowed and the averaged period over whichrepair expenditures were accessed would needto be determined from estimates of the likelyresponses of landlords’ repair expenditures totax incentives.

Any tax breaks introduced would need to be ring-fenced so that the properties remain in theprivately rented sector. This could be done byhaving claw-back provisions. Foregone taxescould be re-imposed if the dwelling is no longermarket-rented, for example, on a declining scalefor, say, a five or ten year period from the grantingof the original tax break.

It is difficult to estimate the full cost of suchmeasures. A rough estimate, nevertheless, ispossible of the gross VAT tax take on repairs. Atpresent, there are around 2 million available tothe public privately rented dwellings in England.Surveys suggest that landlords spend on average£500 a year on repairs. Some will not be chargedVAT, but for simplicity, assume all pay 17.5%VAT on top of the £500. This gives a tax take of£175m a year, though this is then partly offset byrepair costs lowering Schedule A rental incometax revenues.

If repairs and improvements are excessively low inthe privately rented sector, as is often claimed, theleverage effect of tax breaks in inducing additionalexpenditure could be substantial. Deadweight taxlosses to the Treasury on work that would havebeen done in any case would be partly offset bythe extra tax revenue likely from the increase inrental income arising from the existence of betterquality in the privately rented sector.

With so little information to go on, it is hard toevaluate what tax breaks would be required toincrease investment where it is needed. Theargument in Chapter 5 suggested that the repairand improvement deficit might not be as greatas some surveys suggest. It would probably bebest to undertake some pilot experiments toobserve the extent to which landlord investmentwas sensitive to such fiscal stimuli.

3. Housing investment

The big current housing problem in the UK is alack of adequate housing supply in areas wheredemand is high. Incentives for private landlords

66

to invest in stimulating new supply are currentlyabsent. The introduction of PIFs may help butthere is also a need to consider the role that theprivately rented sector can play in stimulatingsupply further. Grants could be offered to buildin localities or in built forms where currentlysupply is insufficient, but not constrained byplanning controls – particularly urbanregeneration areas. Similarly, the blanket use ofaffordable housing criteria in planningobligation requirements to provide new socialhousing could be adjusted to provide moreprivately rented housing as well.

ConclusionA series of reforms have been suggested here,but the overall message is that the growth of theprivately rented sector over the past fifteenyears has been a major success. That successhas primarily been market-driven, yet still it isone in which policy has played an importantpart. Success has come about by allowing a freemarket to operate and the forces of competitionwithin it to work. The UK as a whole is highlyunlikely again to become a nation of privaterenters, but the tenure once more is playing akey and sustainable role in housing provision.The objective of policy should be to keep itthat way.

67

The Social Market Foundation11 Tufton Street

London SW1P 3QB£15.00

ISBN 1 874097 62 3

www.smf.co.uk

The Social Market Foundation is an independent think-tank which explores ideas linking

the state and the market to create the just, free and prosperous Britain of tomorrow.

SMF Private Rental Cover 22/6/04 12:56 am Page 1