Ford’s Five-Dollar Day: Efficiency Wages in Practice
1. Background
On 5 January 1914, the Ford Motor Company announced a minimum wage of $5 for an eight-hour day at its Highland Park plant in Detroit, more than doubling the prevailing rate of roughly $2.34 for a nine-hour shift. The policy was not born of philanthropy. Ford had recently pushed hard into moving-assembly-line methods, compressing chassis assembly from twelve and a half hours to as little as 93 minutes. Output per worker was rising, but the work had become repetitive, tightly paced, and stripped of discretion. The line dictated the rhythm; workers merely followed.
The labour consequences were severe. Annual turnover in 1913 reached 370 to 380 percent: Ford had to hire roughly 52,000 men over the year simply to maintain a workforce of 14,000. Absenteeism ran at around 10 percent daily, and “soldiering” (deliberate slowing of work) was widespread. Recruitment and retraining costs were enormous, and the assembly line’s theoretical efficiency was being undermined by a demoralised, revolving workforce (Raff and Summers, 1987; Meyer, 1981). It was against this backdrop that Henry Ford, together with his vice president James Couzens, devised the Five-Dollar Day. Ford himself later wrote:
“The payment of five dollars a day for an eight-hour day was one of the finest cost-cutting moves we ever made.” (Ford, 1922)
Two institutional details are often missed. First, the $5 was not a simple flat wage: it comprised a base rate of $2.34 plus a “profit-sharing” supplement, conditional on meeting standards of conduct, sobriety, and household management. Second, these conditions were enforced by a newly created Sociological Department that investigated workers’ private lives. The Five-Dollar Day thus combined a higher wage with stronger screening and monitoring.
2. Theoretical Framework
2.1 The Generic Efficiency-Wage Model
Efficiency-wage theory, formalised by Solow (1979) and synthesised in Romer (2019, Ch. 11), holds that the profit-maximising wage need not be the market-clearing wage, because effort is increasing in pay. The representative firm maximises
\pi = F(eL) - wL,
where F(\cdot) is a concave production function, L is employment, w is the wage, and e = e(w) with e'(\cdot) > 0 is the effort function. The first-order condition for L requires
F'(e(w)L) = \frac{w}{e(w)},
so that the marginal product of effective labour equals its unit cost. The first-order condition for w yields the Solow condition (Romer, eq. 11.8):
\frac{w \, e'(w)}{e(w)} = 1.
At the efficiency wage w^*, the elasticity of effort with respect to the wage is unity. The firm minimises the cost per unit of effective labour, w/e(w): a marginal wage increase raises effort in exact proportion, leaving unit cost unchanged. Before 1914, Ford’s assembly line demanded sustained, disciplined effort, yet at $2.34 the observed equilibrium featured low attachment and mass quits. The jump to $5.00 can be read as a discrete move towards the region where effort and retention gains offset the wage premium.
A richer specification (Romer, eq. 11.9) makes effort depend on outside options:
e = e(w, w_a, u), \qquad e_1 > 0, \quad e_2 < 0, \quad e_3 > 0,
where w_a is the wage at competing firms and u is the unemployment rate. Following Summers (1988), with the functional form e = \left(\frac{w - x}{x}\right)^\beta and x = (1 - bu)w_a, imposing the equilibrium condition w = w_a yields
u^{EQ} = \frac{\beta}{b},
determined entirely by the effort-function parameters. Each firm pays a premium of approximately fraction \beta over the index of labour-market conditions, and the incentive to deviate from this wage is small. In January 1914, Detroit’s labour market was tight, making the job-finding rate high and the cost of job loss correspondingly low. The standard wage was insufficient to generate adequate e(w, w_a, u) given assembly-line demands. By setting w = \$5.00 \gg w_a, Ford created a large surplus w - x that raised effort, attracted applicants, and made dismissal genuinely costly to workers.
2.2 The Shapiro–Stiglitz No-Shirking Condition
Shapiro and Stiglitz (1984), presented in Romer §11.2, provide a microfoundation for the effort function through imperfect monitoring. Workers choose between exerting effort \bar{e} (yielding utility w - \bar{e}) and shirking (yielding w, but risking detection at Poisson rate q and firing). The asset-pricing equations on worker value functions are
\rho V_E = (w - \bar{e}) - b(V_E - V_U), \qquad \rho V_S = w - (b + q)(V_S - V_U), \qquad \rho V_U = a(V_E - V_U).
Setting V_E = V_S (the firm pays just enough to deter shirking) and using the steady-state flow balance a = NLb / (\bar{L} - NL), one obtains the no-shirking condition (Romer, eq. 11.37):
w^* = \bar{e} + \left(\rho + \frac{b\bar{L}}{\bar{L} - NL}\right)\frac{\bar{e}}{q}.
Workers must earn a rent over unemployment, V_E - V_U = \bar{e}/q > 0, proportional to the cost of effort and inversely proportional to the detection probability. The NSC slopes upward in employment: as NL \to \bar{L}, dismissed workers find jobs instantly, the penalty of firing vanishes, and no finite wage can deter shirking. Equilibrium occurs where the NSC intersects labour demand, necessarily with involuntary unemployment.
At Ford, assembly work was physically punishing (high \bar{e}), monitoring of sustained attention was imperfect (low q), and the booming Detroit labour market made re-employment easy (high a). All three forces pushed the required wage premium upward. The Sociological Department can be interpreted as an attempt to raise effective q, complementing the wage increase.
3. Empirical Evidence
Raff and Summers (1987) provide the principal empirical study. Using Ford’s internal employment records, they conclude that “the Ford experience strongly supports the relevance of” efficiency-wage theories. The main findings are summarised below.
| Pre-policy (1913) | Post-policy | |
|---|---|---|
| Daily wage | $2.34 (9-hr day) | $5.00 (8-hr day) |
| Annual turnover | 370–380% | 54% (1914); 16% (1915) |
| Daily absenteeism | ~10% | Sharply reduced |
| Cars produced | 202,667 | 308,162 (1914); 501,462 (1915) |
| Applicants at plant gate | n/a | ~10,000 (6 Jan 1914) |
The morning after the announcement, approximately 10,000 applicants gathered outside Highland Park; in subsequent weeks, thousands continued to arrive daily, and Ford had to impose residency requirements to manage the crowd. This queue is the signature prediction of an above-market-clearing wage: if $5 merely compensated for unpleasant conditions, there would have been no excess supply.
Turnover fell from 370 percent in 1913 to 54 percent in 1914 and just 16 percent by 1915 (Slichter, 1921). Ford had previously spent an estimated $35 to $100 per new hire in training costs (Klug, 1989); at over 50,000 hires per year, turnover alone was costing a substantial fraction of the total wage bill. Output per worker rose substantially. Total production increased from 202,667 cars in 1913 to 308,162 in 1914 and 501,462 in 1915, while the workforce grew more modestly. Profits rose from approximately $27 million in 1913 to $32 million in 1914 (a year that also saw a national recession). Raff and Summers (1987, p. S75) report that real profits grew by 15 percent from 1913 to 1914 and by a further 21 percent from 1914 to 1915. The productivity gains, combined with turnover savings, were sufficient to make the Five-Dollar Day profitable for Ford.
Broader evidence supports the mechanism. Krueger and Summers (1988) document persistent interindustry wage differentials: high-wage firms exhibit systematically lower quit rates across industries and time periods, consistent with the rent condition V_E - V_U > 0 functioning as a discipline device. The stability of these differentials across business cycles distinguishes structural efficiency wages from transitory market phenomena.
4. Discussion
Which channels operated?
Multiple efficiency-wage channels operated simultaneously at Ford, and no single mechanism accounts for the full experience. The shirking/effort channel (Shapiro and Stiglitz, 1984) explains the productivity gains and the role of monitoring through the Sociological Department. The turnover channel accounts for the collapse in quits and the associated cost savings. The adverse-selection channel (Weiss, 1980) explains why the conditional structure of the programme improved workforce quality. The morale/fairness channel (Akerlof and Yellen, 1990), supported by later survey evidence (Bewley, 1999; Campbell and Kamlani, 1997), suggests that workers interpreted the above-market wage as fair treatment and reciprocated with higher effort than mere compliance would imply.
Raff and Summers note that the simplest versions of the shirking model are insufficient: many Ford tasks were in fact easily monitored, and the scale of productivity gains suggests that morale and workforce composition effects were at least as important as pure effort incentives.
Identification and competing explanations
A careful evaluation must distinguish “consistent with” from “caused by.” Several confounds deserve attention. First, the assembly line itself was evolving rapidly in 1913 and 1914; substantial output gains would have occurred even without a wage change, making it difficult to attribute the full productivity increase to the Five-Dollar Day alone. Second, Ford changed multiple margins simultaneously: wages, hours, shift organisation (a third shift became possible with the eight-hour day), screening, and monitoring. The policy was a bundle, and isolating the contribution of the wage premium is not straightforward. Third, the popular claim that Ford raised wages so workers could afford Model Ts does not survive scrutiny; the workforce was far too small to absorb the increase in production. The rationale was microeconomic (cost per effective unit of labour), not macroeconomic (aggregate demand).
The episode also had spillovers beyond the firm. Competitors raised wages to retain their own workers, and “Fordism” came to influence broader labour standards and expectations across American industry.
5. Conclusion
Ford’s Five-Dollar Day illustrates efficiency-wage theory in concrete terms. The policy satisfied the Solow condition by pushing w into the region where higher pay was offset by proportionally higher effective labour. It satisfied the no-shirking condition by creating an employment rent, V_E - V_U > 0, large enough to discipline effort on a factory floor where monitoring was imperfect. And it exploited screening and fairness channels through the conditional bonus structure and the Sociological Department. The result was lower unit labour costs, not higher ones.
The broader lesson from the episode, as Romer (2019, Ch. 11) emphasises, is that when labour-market frictions are binding, the profit-maximising wage can lie well above the market-clearing level. The involuntary unemployment that follows (visible in the queues outside Highland Park) is not a market failure to be eliminated, but an equilibrium outcome of firms’ rational compensation decisions.
References
- Akerlof, G.A. and Yellen, J.L. (1990). “The Fair Wage–Effort Hypothesis and Unemployment.” Quarterly Journal of Economics, 105(2), 255–283.
- Bewley, T.F. (1999). Why Wages Don’t Fall during a Recession. Cambridge, MA: Harvard University Press.
- Campbell, C.M. and Kamlani, K.S. (1997). “The Reasons for Wage Rigidity: Evidence from a Survey of Firms.” Quarterly Journal of Economics, 112(3), 759-789.
- Ford, H. (1922). My Life and Work. Garden City, NY: Doubleday.
- Klug, T. (1989). “Employers’ Strategies in the Detroit Labor Market, 1900–1929.” In On the Line: Essays in the History of Auto Work, ed. N. Lichtenstein and S. Meyer. Urbana: University of Illinois Press.
- Krueger, A.B. and Summers, L.H. (1988). “Efficiency Wages and the Inter-Industry Wage Structure.” Econometrica, 56(2), 259–293.
- Meyer, S. (1981). The Five Dollar Day: Labor Management and Social Control in the Ford Motor Company, 1908–1921. Albany: SUNY Press.
- Raff, D.M.G. and Summers, L.H. (1987). “Did Henry Ford Pay Efficiency Wages?” Journal of Labor Economics, 5(4), S57–S86.
- Romer, D. (2019). Advanced Macroeconomics, 5th ed. New York: McGraw-Hill. Chapter 11.
- Shapiro, C. and Stiglitz, J.E. (1984). “Equilibrium Unemployment as a Worker Discipline Device.” American Economic Review, 74(3), 433–444.
- Slichter, S. (1921). The Turnover of Factory Labor. New York: Appleton.
- Solow, R.M. (1979). “Another Possible Source of Wage Stickiness.” Journal of Macroeconomics, 1(1), 79–82.
- Summers, L.H. (1988). “Relative Wages, Efficiency Wages, and Keynesian Unemployment.” American Economic Review, 78(2), 383–388.
- Weiss, A. (1980). “Job Queues and Layoffs in Labor Markets with Flexible Wages.” Journal of Political Economy, 88(3), 526–538.