how does a federal minimum wage hike affect aggregate household spending?

7/28/2019 How does a federal minimum wage hike affect aggregate household spending?

1/4

How does a federal minimum wage hike affect aggregatehousehold spending?by Daniel Aaronson, vice president and director o microeconomic research, and Eric French, senior economist and research advisor

This article finds that a federal minimum wage hike would boost the real income and

spending of minimum wage households. The impact could be sufficient to offset increasing

consumer prices and declining real spending by most non-minimum-wage households

and, therefore, lead to an increase in aggregate household spending. The authors calculate

that a $1.75 hike in the hourly federal minimum wage could increase the level of realgross domestic product (GDP) by up to 0.3 percentage points in the near term, but with

virtually no effect in the long term.

A central part o President Obamas

0 State o the Union address was aproposal to gradually raise the hourly

ederal minimum wage rom $7.5 to $9.Proponents o a higher minimum wage

argue it provides economic stimulus by

putting money intothe hands o people

who are especiallylikely to spend the ex-

tra income. Oppo-nents say a higher

minimum wage orcesrms that employminimum wage work-

ers to cut jobs orraise prices on goods

and services. In this

Chicago Fed Letter, we

use estimates rom ourresearch to analyzeboth arguments.

We begin by assessing the number o

workers whose wages would be aectedby a $.75 hike in the hourly ederal mini-mum wage. Next, based on our prior re-

search, we predict the likely eects o anincrease in the hourly ederal minimum

wage on total household income, con-sumer prices, and aggregate household

spending. We show that a $.75 increase

in the minimum wage could raise realGDP by about 0. percentage points

over the short run (rst year). Allowingmore workers to lose their jobs or allowingthe spending response to be smaller than

our baseline estimates lowers our pro-jected impact o the minimum wage hike

on real GDP over the short run. In addi-tion, we predict the hikes impact on real

GDP to be close to zero over the long run.

We view the minimum wage as essentially

a tax and transer program. Firms thathave to pay higher wages to their work-

ers respond by raising prices on theirgoods and services. Higher prices ongoods and services oset the income

benet or minimum wage workers and

reduce the real income o non-minimum-wage workers who did not get a wageincrease. Still, an increase in aggregate

household spending can arise i minimumwage workers have a higher propensityto spendparticularly in the short run

than non-minimum-wage workers.

Whose wages are affected by aminimum wage hike?

Figure highlights the low end o the

U.S. wage distribution using data rom

Chicago Fed Letter

ESSAYS ON ISSUES THE FEDERAL RESERVE BANK AUGUST 2013

OF CHICAGO NUMBER 313

1. 2012 distribution of wages in U.S. economy

Number of Total wage Share ofWorker workers Share of payments wagecategory (in millions) workers ($ billions) payments$6$7.25/hour 2 0.02 23 0.00

$6$9/hour 15 0.13 204 0.04

$6$10/hour 22 0.19 338 0.07

All hourly workers 69 0.59 2,165 0.43

All workers 117 1.00 5,073 1.00

Notes: Sample weights are used to make the Current Population Survey(CPS) respon-dents comparable to the work force of the U.S. economy aged 16 years and older. Work-ers paid below the minimum wage of $7.25 per hour appear in the CPS mostly on accountof measurement errors in self-reported data. Workers whose reported wages fall below

$6 per hour are excluded. Note that tips are included in the wage payment calculations.source: Authors calculations based on data from the U.S. Bureau of Labor Statistics,Current Population Survey.


2/4

the U.S. Bureau o Labor Statistics

Current Population Survey(CPS). Approx-imately million workers, or % o the

work orce, were paid at or just belowthe current hourly ederal minimum

wage o $7.5 in 0. Roughly 5 mil-lion workers, representing % o the

work orce, made $6$9 per hour (i.e.,at or somewhat below the proposednew ederal minimum). Employers are

not required to raise the wages o work-ers already earning above the new mini-

mum wage. However, in practice theymay. Thereore, we include an addi-tional 7 million workers who made

slightly more than the proposed newederal minimum wagei.e., thoseearning $9$0 per hour.

Although a substantial share o workers

would be aected by minimum wagelegislation, its eect on wage paymentswould be relatively smaller. We estimate

that in 0 roughly $00 billion, or4% o total CPS-reported wage payments,

went to workers earning $6$9 per hour,

and $8 billion, or 7% o total CPS-reported wage payments, went to those

earning $6$0 per hour.

When inerring the likely impact on totalhousehold income, consumer prices,and aggregate household spending

rom the proposed ederal minimumwage hike, we ace two important issues.

First, 9 states and a handul o citiescurrently oer a minimum wage above

and sometimes well abovethe ederalminimum wage. So, i the hourly ederal

minimum wage were raised by $.75,these states and cities might raise theirhourly minimum wages above $9. To

partly account or this, we allow earn-ings and spending to rise somewhat orthe wage group earning $9$0 per hour.

Second, the aggregate wage income o$5.07 trillion computed rom the CPS

or 0 is lower than the aggregatewage income o $6.88 trillion reported

in the U.S. Bureau o Economic Analysiss

National Income and Product Accounts o

the United Statesor that year. Most likely

this dierence arises rom an under-statement o the earnings o high-income

individuals in the CPS, because suchindividuals are dicult to reach viahousehold surveys. I aggregate wage

income has been understated, gure overstates the share o total wage pay-

ments going to low-wage individuals.

Accounting or this possible overstate-ment reduces the share o total wage

payments going to those making $6$9per hour rom 4% to %.

Household income

Next, we compute what happens to totalhousehold income as a result o an in-

crease in the hourly ederal minimumwage rom $7.5 to $9. In Aaronson,

Agarwal, and French (0), we useddata rom three large, representativedata setsthe CPS, the U.S. Census

Bureaus Survey o Income and Program

Participation, and the U.S. Bureau o

Labor Statistics Consumer ExpenditureSurveyto estimate the impact o a min-

imum wage hike on household incomewith adult minimum wage workers. Weound that the average real income o

households with adult minimum wageworkers rose by $50 per quarter dur-

ing the rst ew quarters in response toa $ increase in the minimum wage.4

I we assume that 5 million workersearning $6$9 per hour in 0 receive

a $.75 hourly wage increase and that

the income response is proportional towhat we ound beore, aggregate incomewill rise by $50 $.75 5 million =$6.6 billion per quarter, or roughly

$6 billion during the year immediatelyollowing the hike. Those making

$9$0 per hour likely receive a smallerincome increase than those making less.

Assuming that the income increase orthose earning $9$0 per hour is only one-third o that or those earning $6$9 per

hour (or $50/ = $8), we nd that

those earning $9$0 per hour would

receive $8 $.75 7 million = $ billionper quarter, or $4 billion per year. We

also ound in Aaronson, Agarwal, andFrench (0) that the income responseto a minimum wage increase is isolated

to the groups o workers at and just

above the minimum wage. Thereore,the total income gain or all workers isapproximately $0 billion per year.

Our analysis in Aaronson, Agarwal, andFrench (0) was o adult minimum

wage workersspecically, minimumwage workers who are a households headand spouse aged 8 and older (or in the

absence o a spouse, another workinghousehold member aged at least 8).

Teenagers (unless they happen to becounted as one o their households two

adult workers) and low-skilled workerswithout jobs prior to the minimum wageincrease were omitted rom our analysis.

There is some evidence that minimumwage hikes might make it harder to get

a job, especially or teenagers, who rep-resent % o the minimum wage labororce.5 We return to this issue later.

Consumer prices

Using a variety o U.S. and Canadiandata, we demonstrated in Aaronson

(00) and Aaronson, French, and

MacDonald (008) that immediatelyater a minimum wage increase, limited-service restaurants (i.e., ast-ood restau-rants) employing minimum wage workers

pass close to 00% o the higher laborcosts on to consumers in the orm o

higher prices.

We conjecture that other (nonrestaurant)

rms employing minimum wage workersor using intermediate inputs requiring

minimum wage labor also pass close to00% o the higher labor costs on to

consumers in the orm o higher prices.6

A simple way to predict how a $.75 in-crease in the hourly ederal minimum

wage aects the price level is to comparethe increase in earnings resulting rom

the hike to the level o real GDP (oraggregate prices) or to the level o totalhousehold consumption (or aggregate

consumer prices) under the assumptiono no disemployment eects. Based on

our estimate o a $0 billion earningsimpact in the rst year, we calculate

In the near term, a minimum wage hike can stimulate economic

activity by putting money into the hands of people who are

especially likely to spend it.


3/4

that aggregate prices would rise by

0.9% (= $0 billion/$5.685 trilliono 0 real GDP) and aggregate con-

sumer prices would go up by 0.7%(= $0 billion/$. trillion o 0

total household consumption).7

Aggregate household spendingFinally, to quantiy the aggregate house-hold spending response to a ederalminimum wage hike, we need to con-

sider the spending o both minimumwage and non-minimum-wage earners

in response to the minimum wage hike.

Minimum wage earner spending

In Aaronson, Agarwal, and French(0), we ound that real spending in

households with adult minimum wageworkers rises, on average, by approxi-

mately $700 per quarter during the rstew quarters ollowing a $ hike in thehourly minimum wage. This additional

spending, which exceeds the immedi-ate income gain o $50 per quarter, is

primarily on durable goods, particularlynew vehicles (nanced with credit).

Our research shows that these patternscan be partly reconciled by augmentinga standard dynamic model o consumer

behavior to allow or the ability to bor-row against durable goods. The intuition

or this result is simple. Suppose a house-

hold must make a 0% down paymenton an auto purchase. The existence othis borrowing opportunity implies anextra $50 per quarter in income can

be leveraged up to $,50 ($50/0. =$,50) in additional spending. This

amount o spending is well beyond whatwe nd in the actual data, perhaps be-cause some minimum wage households

cannot nance nondurable purchaseswith credit.

The spending estimate o $700 per quarter

in response to a $ hike in the hourlyminimum wage applies to householdswith adult minimum wage workers. It

seems likely that teenagers, who makeup % o all minimum wage workers,have less access to credit and thereore

will not be able to leverage their earn-ings. Instead, let us assume that teen-

age minimum wage workers spend alltheir income as they earn it. Given the

number o teen and adult workers whoare likely aected by a $.75 hike in the

hourly ederal minimum wage (includingthose earning $9$0 per hour), we cal-

culate that spending among minimumwage households could add as much as

$7 billion to the economy in the yearollowing the hike, which is 0.47% o

real GDP and 0.66% o total household

consumption in 0.

Non-minimum-wage earner spending

Workers who earn above the minimumwage may decrease their real spendingas a consequence o a minimum wage

hike because they typically ace higherproduct and service prices without the

benet o an earnings boost. Supposethat the spending propensity o non-

minimum-wage workers is such that theyreduce their real spending by $800 orevery $,000 o real income lost.8 Those

losing the $,000 o real income throughhigher prices may not reduce their spend-

ing by the ull $,000 but may insteadreduce their savings. We predict that

this loss or non-minimum-wage earnersresults in a $5 billion decline in realspending in the year ollowing the mini-

mum wage hike.

Total spending

Combining the estimates or minimum

wage earners and non-minimum-wageearners, we predict that an increase o

$.75 in the hourly ederal minimumwage raises aggregate household spend-ing by roughly $48 billion in the year

ollowing the minimum wage hike, or0.% o 0 real GDP.

However, a ew words o caution are inorder. First, as we mentioned already,

our analysis is based on household in-come and spending responses romsamples o adult minimum wage work-

ers who had a minimum wage job be-ore the hike. There is some evidence

that minimum wage hikes might makeit harder to get a job, especially or teen-

agers. Additionally, some workers, par-ticularly teenagers, may lose their jobsas a consequence o a minimum wage

hike. For these reasons, we introducedisemployment elasticities o 0.5 or

teenagers and 0.5 or adults (i.e., orevery 0% increase in the minimumwage, the employment o teenagers and

adults making the minimum wage wouldall by 5% and .5%, respectively). Our

reading is that these elasticities are at

the high end o the literature. Never-theless, allowing or disemployment othese magnitudes reduces the aggregate

spending gain ollowing a $.75 hike inthe ederal minimum wage to $8 bil-

lion, or 0.% o 0 real GDP. The

aggregate spending gain would declineto zero i we assume a disemployment

elasticity o 0.7 or both teens andadults. Thereore, while more disem-

ployment than we allow or is certainlyplausible and would clearly lower our

estimate o the spending response, itsunlikely to completely eliminate theentire boost to aggregate spending.

Additionally, or those with low income

and poor credit scores, it may be harderto purchase cars on credit ater the nan-

cial crisis than it was during the sampleperiod o 980008, which we used toestimate the spending response. Indeed,

our estimated aggregate spending re-sponse is high relative to the rest o theliterature. Instead, i we assume that

the marginal propensity to spend (i.e.,the propensity to spend the next dollar)

or households with adult minimumwage workers is hal as large in the year

ollowing a minimum wage hike as what

Charles L. Evans, President;Daniel G. Sullivan,Executive Vice President and Director o Research;Spencer Krane, Senior Vice President and EconomicAdvisor;David Marshall, Senior Vice President, fnancialmarkets group;Daniel Aaronson, Vice President,microeconomic policy research;Jonas D. M. Fisher,Vice President, macroeconomic policy research;RichardHeckinger,Vice President, markets team;Anna L.Paulson, Vice President, fnance team;William A. Testa,Vice President, regional programs, and Economics Editor;Helen OD. Koshy and Han Y. Choi, Editors;Rita Molloy and Julia Baker, Production Editors;Sheila A. Mangler, Editorial Assistant.

Chicago Fed Letteris published by the EconomicResearch Department o the Federal Reserve Banko Chicago. The views expressed are the authorsand do not necessarily refect the views o the

Federal Reserve Bank o Chicago or the FederalReserve System.

0 Federal Reserve Bank o ChicagoChicago Fed Letterarticles may be reproduced in

whole or in part, provided the articles are notreproduced or distributed or commercial gainand provided the source is appropriately credited.Prior written permission must be obtained orany other reproduction, distribution, republica-tion, or creation o derivative works oChicago FedLetterarticles. To request permission, please contactHelen Koshy, senior editor, at --580 oremail [email protected]. Chicago FedLetterand other Bank publications are availableat www.chicagoed.org.ISSN 0895-064


4/4

we estimate in the data, the aggregate

spending response to a $.75 increasein the hourly ederal minimum wage

would be only $4 billion, or 0.0% o0 real GDP. This result highlights

the mechanism o our predictionanyadditional consumer spending rom a

minimum wage hike arises rom di-erences in the propensity to spendamong dierent income groups.

Finally, its important to stress that the

aggregate household spending responsediscussed in this article is relevant or

only the rst ew quarters ater a mini-mum wage hike. Beyond that time rame,households must pay o debt they in-

curred in the short run by spending less.

Thus, a minimum wage hike providesstimulus or a year or so, but serves as adrag on the economy beyond that.

Conclusion

Proponents o minimum wage increasesoten claim that minimum wage hikes

will signicantly boost the economy. Weare skeptical that minimum wage hikes

boost GDP in the long run. Nevertheless,we do nd evidence that putting money

into the hands o consumers, especiallylow-wage consumers, leads to predictableincreases in spending in the short run.

1 See, e.g., New York Times Company, 0,From the bottom up, New York Times,February 7, available at www.nytimes.com/0/0/8/opinion/wages-rom-the-bottom-up.html.

2 Daniel Aaronson, Sumit Agarwal, andEric French, 0, The spending anddebt response to minimum wage hikes,American Economic Review, Vol. 0, No. 7,December, pp. 9; Daniel Aaronson,

Eric French, and James MacDonald, 008,The minimum wage, restaurant prices,and labor market structure,Journal oHuman Resources, Vol. 4, No. , Summer,pp. 68870; and Daniel Aaronson, 00,Price pass-through and the minimum wage,Review o Economics and Statistics, Vol. 8,No. , February, pp. 5869.

3 For urther explanation o the calculationsin this article, see www.chicagoed.org/digital_assets/others/people/research_resources/aaronson_daniel/aaronson_rench_cf__calculations.xlsx andwww.chicagoed.org/digital_assets/others/people/research_resources/aaronson_daniel/aaronson_rench_cf__calculations_documentation.pd.

4To put this estimate into perspective, note

that adult minimum wage employees work,on average, roughly 00 hours per quarter.Under three assumptionsthere is no dis-employment (i.e., job loss) due to the min-imum wage hike, all workers who are paidclose to the minimum wage are coveredby minimum wage laws, and there is nomeasurement errorwe anticipate a $minimum wage hike to increase eachadult minimum wage employees quarterlyearnings by $00.

5U.S. Bureau o Labor Statistics, 0, Char-acteristics o minimum wage workers: 0,report, Washington, DC, March , availableat www.bls.gov/cps/minwage0.htm.

6Prices and incomes might also rise ollow-ing a minimum wage hike because o theincrease in aggregate demand or goodsand services. We do not account or thispossibility in our analysis.

7Real GDP and total household consump-tion data are rom the U.S. Bureau oEconomic Analysis.

8See, e.g., Jonathan A. Parker, Nicholas S.Souleles, David S. Johnson, and RobertMcClelland, 0, Consumer spendingand the economic stimulus payments o008, National Bureau o EconomicResearch, working paper, No. 6684,January, available at www.nber.org/papers/w6684.

how does a federal minimum wage hike affect aggregate household spending?

Documents