Efficiency wages

An efficiency wage is a wage above the market-clearing level that a profit-maximizing firm pays because the higher wage raises productivity by more than it raises cost. Because firms do not cut pay when workers queue for their jobs, efficiency wages produce involuntary unemployment and persistent wage differences between equally skilled workers.

Why pay more than the market wage

A higher wage can lower the cost of a unit of effective labour in four ways.

1. Shirking. A well-paid job is costly to lose, so the threat of firing deters shirking.¹
2. Turnover. Workers quit less, saving hiring and training costs.²
3. Adverse selection. A higher wage attracts a better pool of applicants.³
4. Fairness. Workers who feel well treated reciprocate with effort: partial gift exchange.⁴ In market experiments, buyers offered prices well above the market-clearing level and sellers responded, on average, with higher quality.⁵

Example: a firm that pays $15 when the going rate is $14 gains if the extra dollar buys enough less shirking, fewer quits or better applicants to more than pay for itself.

The Solow condition

Let effort per hour depend on the wage, $e(w)$. The firm buys effort, so it minimizes the cost of a unit of effort, $w/e(w)$. The derivative is $(e - w\,e')/e^2$; setting it to zero gives the Solow condition: the elasticity of effort with respect to the wage equals one,

$$\frac{de}{dw}\cdot\frac{w}{e} = 1.$$

Wage paid (w)

$20.00

Effort e(w)

10.00

Effort per dollar e/w

0.500

Cost per unit of effort w/e

$2.00

Elasticity of effort with respect to the wage

1.00

Open in Graph Lab

Figure 1. With effort $e(w) = \sqrt{10(w - 10)}$, the ray from the origin is steepest, at 0.5 units of effort per dollar, where it is tangent to the effort curve at $w = 20$. Any other wage gives a flatter ray and a higher cost per unit of effort than $2.

The efficiency wage depends only on the effort function, not on labour supply.⁶ When demand falls, the firm keeps the wage and cuts jobs, which is one reason wages adjust slowly in recessions.

The shirking model

Shapiro and Stiglitz model identical workers who either work or shirk.¹ A worker who works bears an effort cost $e$, measured in dollars per period. A shirker saves $e$ but is caught at rate $q$ and fired. Jobs also end for other reasons at rate $b$, the interest rate is $r$, and an unemployed worker receives $\bar{w}$. With unemployment rate $u$, the wage must satisfy the no-shirking condition (NSC):

$$w \ge \bar{w} + e + \frac{e}{q}\left(r + \frac{b}{u}\right).$$

A shirker gains $e$ each period but loses the job’s value over unemployment at rate $q$, so working pays when that value is at least $e/q$; the NSC is the wage that delivers it. The wage covers the benefit $\bar{w}$ and the effort cost $e$, plus a premium that makes the job worth keeping. The premium is larger when shirkers are rarely caught (low $q$) and when a fired worker would quickly find another job. In a steady state the rate at which the unemployed find jobs plus $b$ equals $b/u$, so the premium rises without limit as $u$ approaches zero.

Key idea

At full employment a fired shirker walks straight into another job, so no wage can deter shirking. Unemployment is the worker discipline device: in equilibrium some workers who would work for less cannot find jobs.

Equilibrium is where labour demand meets the NSC. Supply is vertical at the labour force; if effort were observable, the market would clear there.

Detection rate q (chance per year a shirker is caught)

1.0

Unemployment benefit (dollars per hour)

$8.00

Efficiency wage

$15.49

Employment

92.6k

Unemployment rate

7.4%

Market-clearing wage if effort were observable

$14.00

Open in Graph Lab

Figure 2. With a detection rate of 1 a year and an $8 benefit, labour demand meets the NSC at a wage of $15.49 and employment of 92.6 thousand, so 7.4 percent of the labour force is unemployed; with observable effort the market would clear at $14 with full employment. Better monitoring lowers the NSC and unemployment; a higher benefit raises both.

The unemployment is involuntary. An unemployed worker who offers to work for less than $w$ is not hired, because at a lower wage the firm expects that worker to shirk. Three predictions follow:

• Monitoring substitutes for wages. A higher $q$ lowers the NSC: lower wages, more employment.
• Unemployment benefits raise unemployment. A higher $\bar{w}$ makes being fired less painful, raises the NSC and reduces employment.
• Demand shocks move employment more than wages. A fall in demand slides the equilibrium down the steep NSC, raising unemployment with only a small wage cut.

Worked example

Effort costs $2 an hour ($e = 2$), $\bar{w} = 8$, $q = 1$, $b = 0.2$ and $r = 0.05$ per year. Labour demand is $w = 34 - 0.2E$ and the labour force is 100 thousand. These are the values in Figure 2.

1. NSC. $w = 8 + 2 + 2(0.05 + 0.2/u) = 10.1 + 0.4/u$.

2. Demand in terms of $u$. Employment is $E = 100(1 - u)$, so $w = 34 - 20(1 - u) = 14 + 20u$.

3. Solve. $14 + 20u = 10.1 + 0.4/u$. Multiply by $u$: $20u^2 + 3.9u - 0.4 = 0$, so $u = \dfrac{-3.9 + \sqrt{3.9^2 + 32}}{40} = 0.0743$.

4. Read off the equilibrium. Unemployment is 7.4 percent, $E = 92.6$ thousand and $w = 14 + 20(0.0743) = 15.49$.

5. Benchmark. With observable effort the market clears at full employment: $w = 34 - 0.2(100) = 14$.

6. Double monitoring. With $q = 2$ the NSC becomes $w = 10.05 + 0.2/u$. Solving $20u^2 + 3.95u - 0.2 = 0$ gives $u = 4.2$ percent and $w = 14.84$.

Exercise: the Solow wage

With $e(w) = \sqrt{10(w - 10)}$, $de/dw = 5/\sqrt{10(w - 10)}$, so the elasticity is $\dfrac{5w}{10(w - 10)} = \dfrac{w}{2(w - 10)}$. Setting it to 1 gives $w = 2(w - 10)$, so $w = 20$. Effort is $\sqrt{100} = 10$ and the cost of a unit of effort is $20/10 = 2$ dollars. At $15 or $30 effort per dollar is 0.471, so a unit of effort costs $2.12.

Ford’s five-dollar day

On January 5, 1914, Ford announced that from January 12 it would cut the working day from 9 to 8 hours and raise minimum daily pay from about $2.34 to $5.00 for qualifying workers: men over 22 whom the company’s Sociological Department judged fit to receive it, with a six-month service requirement added that autumn. The extra pay was labelled profit sharing.⁷

The problem it addressed was severe. In 1913 turnover reached 370 percent: Ford hired 50,448 men to keep an average workforce of 13,623, and 10 percent of workers were absent on a typical day. After the five-dollar day:

• turnover fell to 54 percent in 1914 and 16 percent in 1915;
• absenteeism fell from 10 percent (October 1913) to 2.5 percent (October 1914);
• on the day the plan took effect, twelve thousand job seekers rushed the plant gates, and queues persisted for weeks;
• regressions on plant data suggest productivity rose 40 to 70 percent, and profits rose in 1914 and 1915.⁷

Raff and Summers conclude that the episode strongly supports efficiency wage theory. The raise answered labour problems of the kind the theory stresses, the queues show the wage was far above what Ford needed to attract workers, and productivity and profits rose. Lower turnover costs offset only about 6 percent of the program’s cost; the gains came mainly from harder work, with absenteeism and discharges falling sharply.⁷

Evidence: wage differences across industries

The model predicts that equally skilled workers earn more in industries where effort is costly to monitor or turnover is expensive. Krueger and Summers found substantial differences in wages across US industries even after allowing for measured and unmeasured worker quality, working conditions, fringe benefits and union threats, and found lower turnover in high-wage industries.⁸ Canadian data from the 1986 Labour Market Activity Survey show industry differentials that are substantial, stable over time, similar across kinds of workers and hard to explain by compensating factors.⁹ Workers who lose jobs and switch industries see wage changes that closely match the industry differentials estimated in a cross-section, which is hard to square with pure unmeasured ability.¹⁰

Direct tests point the same way:

• Across one firm’s plants, a higher wage premium went with fewer dismissals for shirking.¹¹
• Company-owned fast-food outlets, where the managers who supervise workers are employees rather than owners, paid 9 percent more to assistant and shift managers and 2 percent more to full-time crew than franchised outlets.¹²
• Jobs in high-wage industries, large firms and highly unionized firms attracted relatively many applicants: queues.¹³

Unemployment rate, Canada

Monthly, seasonally adjusted, population aged 15 and over

Canada's unemployment rate has not fallen below 4.8 percent in any month since 1976, its low in July 2022. Job search explains much of this floor; the shirking model adds a reason why even a booming market keeps some unemployment.Source: Statistics Canada, Table 14-10-0287-01, Labour force characteristics, monthly, seasonally adjusted and trend-cycle. Retrieved 2026-10-23.

Show the data and notes

Seasonally adjusted Labour Force Survey estimates, both genders; revised periodically. The Labour Force Survey covers the 10 provinces; it excludes the territories, people living on reserves, full-time members of the Canadian Armed Forces and people living in institutions.

Period	Unemployment rate
Jan 1976	7.1%
Feb 1976	7.0%
Mar 1976	6.7%
Apr 1976	6.8%
May 1976	6.9%
Jun 1976	6.9%
Jul 1976	7.4%
Aug 1976	7.1%
Sep 1976	7.0%
Oct 1976	7.4%
Nov 1976	7.4%
Dec 1976	7.5%
Jan 1977	7.6%
Feb 1977	7.9%
Mar 1977	7.8%
Apr 1977	7.9%
May 1977	7.8%
Jun 1977	7.8%
Jul 1977	8.1%
Aug 1977	8.2%
Sep 1977	8.3%
Oct 1977	8.4%
Nov 1977	8.5%
Dec 1977	8.5%
Jan 1978	8.3%
Feb 1978	8.3%
Mar 1978	8.5%
Apr 1978	8.4%
May 1978	8.6%
Jun 1978	8.4%
Jul 1978	8.3%
Aug 1978	8.4%
Sep 1978	8.4%
Oct 1978	8.2%
Nov 1978	8.3%
Dec 1978	8.3%
Jan 1979	8.2%
Feb 1979	8.0%
Mar 1979	7.9%
Apr 1979	8.0%
May 1979	7.6%
Jun 1979	7.4%
Jul 1979	7.2%
Aug 1979	7.1%
Sep 1979	7.0%
Oct 1979	7.2%
Nov 1979	7.2%
Dec 1979	7.2%
Jan 1980	7.5%
Feb 1980	7.6%
Mar 1980	7.6%
Apr 1980	7.7%
May 1980	7.8%
Jun 1980	7.7%
Jul 1980	7.6%
Aug 1980	7.6%
Sep 1980	7.3%
Oct 1980	7.3%
Nov 1980	7.2%
Dec 1980	7.3%
Jan 1981	7.4%
Feb 1981	7.4%
Mar 1981	7.4%
Apr 1981	7.1%
May 1981	7.2%
Jun 1981	7.2%
Jul 1981	7.2%
Aug 1981	7.1%
Sep 1981	8.1%
Oct 1981	8.3%
Nov 1981	8.3%
Dec 1981	8.7%
Jan 1982	8.6%
Feb 1982	8.9%
Mar 1982	9.3%
Apr 1982	9.8%
May 1982	10.3%
Jun 1982	11.1%
Jul 1982	11.9%
Aug 1982	12.0%
Sep 1982	12.4%
Oct 1982	12.9%
Nov 1982	12.9%
Dec 1982	13.1%
Jan 1983	12.7%
Feb 1983	12.7%
Mar 1983	12.5%
Apr 1983	12.4%
May 1983	12.4%
Jun 1983	12.4%
Jul 1983	11.9%
Aug 1983	11.7%
Sep 1983	11.4%
Oct 1983	11.3%
Nov 1983	11.3%
Dec 1983	11.3%
Jan 1984	11.3%
Feb 1984	11.3%
Mar 1984	11.3%
Apr 1984	11.5%
May 1984	11.7%
Jun 1984	11.3%
Jul 1984	11.2%
Aug 1984	11.3%
Sep 1984	11.8%
Oct 1984	11.3%
Nov 1984	11.4%
Dec 1984	11.1%
Jan 1985	10.6%
Feb 1985	10.8%
Mar 1985	11.0%
Apr 1985	10.8%
May 1985	10.6%
Jun 1985	10.7%
Jul 1985	10.4%
Aug 1985	10.3%
Sep 1985	10.2%
Oct 1985	10.3%
Nov 1985	10.3%
Dec 1985	10.1%
Jan 1986	9.8%
Feb 1986	9.9%
Mar 1986	9.8%
Apr 1986	9.7%
May 1986	9.5%
Jun 1986	9.6%
Jul 1986	9.6%
Aug 1986	9.6%
Sep 1986	9.5%
Oct 1986	9.4%
Nov 1986	9.4%
Dec 1986	9.5%
Jan 1987	9.5%
Feb 1987	9.5%
Mar 1987	9.4%
Apr 1987	9.2%
May 1987	8.9%
Jun 1987	8.9%
Jul 1987	8.7%
Aug 1987	8.6%
Sep 1987	8.4%
Oct 1987	8.3%
Nov 1987	8.2%
Dec 1987	8.0%
Jan 1988	8.1%
Feb 1988	7.8%
Mar 1988	7.8%
Apr 1988	7.7%
May 1988	7.8%
Jun 1988	7.6%
Jul 1988	7.8%
Aug 1988	7.8%
Sep 1988	7.8%
Oct 1988	7.8%
Nov 1988	7.8%
Dec 1988	7.5%
Jan 1989	7.5%
Feb 1989	7.6%
Mar 1989	7.5%
Apr 1989	7.8%
May 1989	7.7%
Jun 1989	7.5%
Jul 1989	7.5%
Aug 1989	7.3%
Sep 1989	7.3%
Oct 1989	7.2%
Nov 1989	7.5%
Dec 1989	7.7%
Jan 1990	7.9%
Feb 1990	7.7%
Mar 1990	7.3%
Apr 1990	7.6%
May 1990	7.8%
Jun 1990	7.6%
Jul 1990	7.9%
Aug 1990	8.1%
Sep 1990	8.5%
Oct 1990	8.8%
Nov 1990	9.1%
Dec 1990	9.5%
Jan 1991	9.8%
Feb 1991	10.2%
Mar 1991	10.5%
Apr 1991	10.3%
May 1991	10.2%
Jun 1991	10.5%
Jul 1991	10.5%
Aug 1991	10.5%
Sep 1991	10.3%
Oct 1991	10.3%
Nov 1991	10.4%
Dec 1991	10.3%
Jan 1992	10.4%
Feb 1992	10.5%
Mar 1992	10.9%
Apr 1992	10.7%
May 1992	10.9%
Jun 1992	11.4%
Jul 1992	11.3%
Aug 1992	11.7%
Sep 1992	11.6%
Oct 1992	11.4%
Nov 1992	12.1%
Dec 1992	11.7%
Jan 1993	11.2%
Feb 1993	11.0%
Mar 1993	11.2%
Apr 1993	11.6%
May 1993	11.6%
Jun 1993	11.7%
Jul 1993	11.6%
Aug 1993	11.2%
Sep 1993	11.5%
Oct 1993	11.3%
Nov 1993	11.2%
Dec 1993	11.4%
Jan 1994	11.4%
Feb 1994	11.1%
Mar 1994	10.6%
Apr 1994	10.9%
May 1994	10.7%
Jun 1994	10.3%
Jul 1994	10.1%
Aug 1994	10.2%
Sep 1994	10.1%
Oct 1994	10.0%
Nov 1994	9.7%
Dec 1994	9.6%
Jan 1995	9.6%
Feb 1995	9.6%
Mar 1995	9.7%
Apr 1995	9.5%
May 1995	9.5%
Jun 1995	9.5%
Jul 1995	9.6%
Aug 1995	9.5%
Sep 1995	9.2%
Oct 1995	9.3%
Nov 1995	9.2%
Dec 1995	9.4%
Jan 1996	9.4%
Feb 1996	9.5%
Mar 1996	9.6%
Apr 1996	9.3%
May 1996	9.2%
Jun 1996	9.8%
Jul 1996	9.7%
Aug 1996	9.4%
Sep 1996	9.9%
Oct 1996	9.9%
Nov 1996	9.9%
Dec 1996	9.7%
Jan 1997	9.5%
Feb 1997	9.5%
Mar 1997	9.3%
Apr 1997	9.4%
May 1997	9.4%
Jun 1997	9.1%
Jul 1997	8.9%
Aug 1997	8.9%
Sep 1997	8.8%
Oct 1997	8.9%
Nov 1997	8.9%
Dec 1997	8.5%
Jan 1998	8.8%
Feb 1998	8.6%
Mar 1998	8.4%
Apr 1998	8.3%
May 1998	8.3%
Jun 1998	8.4%
Jul 1998	8.3%
Aug 1998	8.1%
Sep 1998	8.2%
Oct 1998	8.0%
Nov 1998	8.0%
Dec 1998	8.1%
Jan 1999	7.9%
Feb 1999	7.9%
Mar 1999	7.9%
Apr 1999	8.2%
May 1999	7.9%
Jun 1999	7.6%
Jul 1999	7.6%
Aug 1999	7.4%
Sep 1999	7.5%
Oct 1999	7.2%
Nov 1999	6.9%
Dec 1999	6.8%
Jan 2000	6.8%
Feb 2000	6.9%
Mar 2000	6.9%
Apr 2000	6.7%
May 2000	6.6%
Jun 2000	6.7%
Jul 2000	6.8%
Aug 2000	7.0%
Sep 2000	6.9%
Oct 2000	7.0%
Nov 2000	6.9%
Dec 2000	6.8%
Jan 2001	6.9%
Feb 2001	7.0%
Mar 2001	7.1%
Apr 2001	7.1%
May 2001	7.0%
Jun 2001	7.2%
Jul 2001	7.1%
Aug 2001	7.2%
Sep 2001	7.2%
Oct 2001	7.3%
Nov 2001	7.5%
Dec 2001	8.1%
Jan 2002	8.0%
Feb 2002	7.9%
Mar 2002	7.9%
Apr 2002	7.7%
May 2002	7.8%
Jun 2002	7.6%
Jul 2002	7.6%
Aug 2002	7.4%
Sep 2002	7.6%
Oct 2002	7.6%
Nov 2002	7.5%
Dec 2002	7.6%
Jan 2003	7.5%
Feb 2003	7.5%
Mar 2003	7.4%
Apr 2003	7.6%
May 2003	7.8%
Jun 2003	7.6%
Jul 2003	7.7%
Aug 2003	7.8%
Sep 2003	7.9%
Oct 2003	7.6%
Nov 2003	7.4%
Dec 2003	7.3%
Jan 2004	7.3%
Feb 2004	7.3%
Mar 2004	7.3%
Apr 2004	7.2%
May 2004	7.1%
Jun 2004	7.2%
Jul 2004	7.1%
Aug 2004	7.0%
Sep 2004	6.9%
Oct 2004	7.1%
Nov 2004	7.2%
Dec 2004	7.1%
Jan 2005	6.9%
Feb 2005	7.0%
Mar 2005	6.9%
Apr 2005	6.7%
May 2005	7.0%
Jun 2005	6.8%
Jul 2005	6.7%
Aug 2005	6.7%
Sep 2005	6.7%
Oct 2005	6.7%
Nov 2005	6.3%
Dec 2005	6.6%
Jan 2006	6.7%
Feb 2006	6.6%
Mar 2006	6.5%
Apr 2006	6.5%
May 2006	6.2%
Jun 2006	6.3%
Jul 2006	6.5%
Aug 2006	6.5%
Sep 2006	6.5%
Oct 2006	6.3%
Nov 2006	6.5%
Dec 2006	6.3%
Jan 2007	6.4%
Feb 2007	6.3%
Mar 2007	6.3%
Apr 2007	6.3%
May 2007	6.2%
Jun 2007	6.1%
Jul 2007	6.0%
Aug 2007	6.0%
Sep 2007	6.0%
Oct 2007	6.0%
Nov 2007	6.1%
Dec 2007	6.2%
Jan 2008	6.1%
Feb 2008	6.1%
Mar 2008	6.2%
Apr 2008	6.2%
May 2008	6.2%
Jun 2008	6.1%
Jul 2008	6.2%
Aug 2008	6.2%
Sep 2008	6.3%
Oct 2008	6.4%
Nov 2008	6.7%
Dec 2008	7.0%
Jan 2009	7.5%
Feb 2009	8.1%
Mar 2009	8.4%
Apr 2009	8.4%
May 2009	8.7%
Jun 2009	8.8%
Jul 2009	8.8%
Aug 2009	8.7%
Sep 2009	8.5%
Oct 2009	8.6%
Nov 2009	8.6%
Dec 2009	8.6%
Jan 2010	8.4%
Feb 2010	8.4%
Mar 2010	8.3%
Apr 2010	8.2%
May 2010	8.2%
Jun 2010	8.0%
Jul 2010	8.2%
Aug 2010	8.2%
Sep 2010	8.2%
Oct 2010	8.1%
Nov 2010	7.9%
Dec 2010	7.8%
Jan 2011	7.8%
Feb 2011	7.8%
Mar 2011	7.8%
Apr 2011	7.8%
May 2011	7.7%
Jun 2011	7.7%
Jul 2011	7.4%
Aug 2011	7.4%
Sep 2011	7.5%
Oct 2011	7.5%
Nov 2011	7.6%
Dec 2011	7.5%
Jan 2012	7.7%
Feb 2012	7.6%
Mar 2012	7.4%
Apr 2012	7.4%
May 2012	7.6%
Jun 2012	7.4%
Jul 2012	7.3%
Aug 2012	7.4%
Sep 2012	7.4%
Oct 2012	7.4%
Nov 2012	7.4%
Dec 2012	7.3%
Jan 2013	7.1%
Feb 2013	7.1%
Mar 2013	7.3%
Apr 2013	7.2%
May 2013	7.1%
Jun 2013	7.2%
Jul 2013	7.3%
Aug 2013	7.3%
Sep 2013	7.1%
Oct 2013	7.3%
Nov 2013	7.1%
Dec 2013	7.4%
Jan 2014	7.3%
Feb 2014	7.3%
Mar 2014	7.1%
Apr 2014	7.1%
May 2014	7.3%
Jun 2014	7.2%
Jul 2014	7.1%
Aug 2014	7.0%
Sep 2014	7.0%
Oct 2014	6.8%
Nov 2014	6.8%
Dec 2014	6.7%
Jan 2015	6.8%
Feb 2015	6.9%
Mar 2015	6.9%
Apr 2015	7.0%
May 2015	6.9%
Jun 2015	6.9%
Jul 2015	7.0%
Aug 2015	7.1%
Sep 2015	7.1%
Oct 2015	7.0%
Nov 2015	7.1%
Dec 2015	7.3%
Jan 2016	7.2%
Feb 2016	7.3%
Mar 2016	7.2%
Apr 2016	7.3%
May 2016	7.1%
Jun 2016	7.0%
Jul 2016	7.0%
Aug 2016	6.9%
Sep 2016	7.0%
Oct 2016	6.9%
Nov 2016	6.8%
Dec 2016	6.9%
Jan 2017	6.9%
Feb 2017	6.6%
Mar 2017	6.7%
Apr 2017	6.5%
May 2017	6.6%
Jun 2017	6.5%
Jul 2017	6.3%
Aug 2017	6.2%
Sep 2017	6.2%
Oct 2017	6.3%
Nov 2017	6.2%
Dec 2017	6.0%
Jan 2018	5.9%
Feb 2018	6.0%
Mar 2018	5.8%
Apr 2018	5.8%
May 2018	5.9%
Jun 2018	6.0%
Jul 2018	5.9%
Aug 2018	6.0%
Sep 2018	5.8%
Oct 2018	5.6%
Nov 2018	5.7%
Dec 2018	5.6%
Jan 2019	5.7%
Feb 2019	5.8%
Mar 2019	5.8%
Apr 2019	5.7%
May 2019	5.5%
Jun 2019	5.6%
Jul 2019	5.7%
Aug 2019	5.7%
Sep 2019	5.6%
Oct 2019	5.6%
Nov 2019	5.9%
Dec 2019	5.6%
Jan 2020	5.5%
Feb 2020	5.8%
Mar 2020	8.5%
Apr 2020	13.7%
May 2020	14.2%
Jun 2020	12.6%
Jul 2020	10.9%
Aug 2020	10.1%
Sep 2020	9.1%
Oct 2020	9.0%
Nov 2020	8.6%
Dec 2020	8.9%
Jan 2021	9.2%
Feb 2021	8.5%
Mar 2021	7.7%
Apr 2021	8.2%
May 2021	8.3%
Jun 2021	7.9%
Jul 2021	7.4%
Aug 2021	7.1%
Sep 2021	7.0%
Oct 2021	6.5%
Nov 2021	6.1%
Dec 2021	5.9%
Jan 2022	6.4%
Feb 2022	5.5%
Mar 2022	5.4%
Apr 2022	5.4%
May 2022	5.2%
Jun 2022	4.9%
Jul 2022	4.8%
Aug 2022	5.2%
Sep 2022	5.1%
Oct 2022	5.1%
Nov 2022	5.0%
Dec 2022	5.0%
Jan 2023	5.1%
Feb 2023	5.2%
Mar 2023	5.1%
Apr 2023	5.1%
May 2023	5.2%
Jun 2023	5.4%
Jul 2023	5.4%
Aug 2023	5.4%
Sep 2023	5.5%
Oct 2023	5.7%
Nov 2023	5.8%
Dec 2023	5.8%
Jan 2024	5.7%
Feb 2024	5.9%
Mar 2024	6.1%
Apr 2024	6.2%
May 2024	6.3%
Jun 2024	6.4%
Jul 2024	6.4%
Aug 2024	6.6%
Sep 2024	6.5%
Oct 2024	6.6%
Nov 2024	7.0%
Dec 2024	6.7%
Jan 2025	6.7%
Feb 2025	6.6%
Mar 2025	6.8%
Apr 2025	6.9%
May 2025	7.0%
Jun 2025	6.9%
Jul 2025	6.9%
Aug 2025	7.1%
Sep 2025	7.1%
Oct 2025	6.9%
Nov 2025	6.6%
Dec 2025	6.8%
Jan 2026	6.5%
Feb 2026	6.7%
Mar 2026	6.7%
Apr 2026	6.9%
May 2026	6.6%
Jun 2026	6.5%
Jul 2026	6.4%
Aug 2026	6.4%
Sep 2026	6.5%

Common mistakes

Traps on exams

• Calling an efficiency wage a compensating differential. Workers are identical and the jobs need not be unpleasant; the premium buys effort, not compensation for a disamenity.
• Calling the unemployed voluntarily idle. They would work at the going wage or less, but firms will not hire them at a lower wage because it invites shirking.
• Drawing equilibrium at full employment. The NSC rises without limit as unemployment approaches zero, so it never meets demand at the labour force.
• Assuming higher pay lowers profit. The firm chooses the wage that minimizes the cost of effort; paying less would raise it.

On the exam

Draw employment on the horizontal axis with a vertical labour supply at the labour force, the downward-sloping demand curve and an NSC that rises steeply as employment approaches the labour force. Mark the efficiency wage where demand meets the NSC and label the gap to the labour force as unemployment.

Check yourself

1.

In the Shapiro-Stiglitz shirking model, why can the economy never reach full employment?

1At full employment a fired shirker would immediately find another job, so no wage could make the threat of firing deter shirking2Unemployment benefits are too generous for anyone to work at full employment3Firms cannot afford to hire every worker at the market-clearing wage4Some workers prefer leisure at any wage

Check answer

2.

In the shirking model, the government raises unemployment benefits. What happens?

1The no-shirking condition shifts up, so the wage rises and employment falls2The no-shirking condition shifts down, because workers no longer need high wages3Wages and employment are unchanged, because benefits only affect the unemployed4Employment rises, because workers search longer for better jobs

Check answer

3.

Which evidence did Raff and Summers find after Ford introduced the five-dollar day in 1914?

1Turnover and absenteeism fell sharply, long queues formed for Ford jobs, and productivity and profits rose2Turnover rose because workers took the higher pay and quit sooner3Ford could not fill its jobs until it raised pay further4Productivity fell, so the higher wage was paid out of profits

Check answer

Practise more questions on this topic

Sources

1. Shapiro, C. and J. E. Stiglitz (1984). Equilibrium unemployment as a worker discipline device. American Economic Review, 74(3), 433 to 444. IDEAS. ↩ ↩²

2. Salop, S. C. (1979). A model of the natural rate of unemployment. American Economic Review, 69(1), 117 to 125. IDEAS. ↩

3. Weiss, A. (1980). Job queues and layoffs in labor markets with flexible wages. Journal of Political Economy, 88(3), 526 to 538. doi:10.1086/260884. ↩

4. Akerlof, G. A. (1982). Labor contracts as partial gift exchange. Quarterly Journal of Economics, 97(4), 543 to 569. doi:10.2307/1885099. ↩

5. Fehr, E., G. Kirchsteiger and A. Riedl (1993). Does fairness prevent market clearing? An experimental investigation. Quarterly Journal of Economics, 108(2), 437 to 459. doi:10.2307/2118338. ↩

6. Solow, R. M. (1979). Another possible source of wage stickiness. Journal of Macroeconomics, 1(1), 79 to 82. doi:10.1016/0164-0704(79)90022-3. ↩

7. Raff, D. M. G. and L. H. Summers (1987). Did Henry Ford pay efficiency wages? Journal of Labor Economics, 5(4, part 2), S57 to S86. NBER working paper 2101. ↩ ↩² ↩³

8. Krueger, A. B. and L. H. Summers (1988). Efficiency wages and the inter-industry wage structure. Econometrica, 56(2), 259 to 293. doi:10.2307/1911072. ↩

9. Gera, S. and G. Grenier (1994). Interindustry wage differentials and efficiency wages: Some Canadian evidence. Canadian Journal of Economics, 27(1), 81 to 100. IDEAS. ↩

10. Gibbons, R. and L. F. Katz (1992). Does unmeasured ability explain inter-industry wage differentials? Review of Economic Studies, 59(3), 515 to 535. doi:10.2307/2297862. ↩

11. Cappelli, P. and K. Chauvin (1991). An interplant test of the efficiency wage hypothesis. Quarterly Journal of Economics, 106(3), 769 to 787. doi:10.2307/2937926. ↩

12. Krueger, A. B. (1991). Ownership, agency, and wages: An examination of franchising in the fast food industry. Quarterly Journal of Economics, 106(1), 75 to 101. doi:10.2307/2937907. ↩

13. Holzer, H. J., L. F. Katz and A. B. Krueger (1991). Job queues and wages. Quarterly Journal of Economics, 106(3), 739 to 768. doi:10.2307/2937925. ↩