Richard Lester's Institutional‐Industrial Relations Model of Labor Markets and the Near‐Zero Minimum Wage Employment Effect: The Model Card and Krueger Ignored but Shouldn't Have

Abstract:

David Card and Alan Krueger dedicate their minimum wage book Myth and Measurement (1995) to Richard Lester, an institutional‐industrial relations labor economist and key figure in the marginalist controversy of the 1940s. Lester claimed a minimum wage law's employment effect is likely zero or near‐zero, the same as Card and Krueger found a half‐century later, but they did not follow Lester's theoretical explanation and instead advanced a marginalist dynamic monopsony model. Numerous empirical studies have followed, with substantial evidence pointing to a relatively small and perhaps zero employment effect, but existing theoretical models, such as competitive, search, and monopsony, remain unable to provide a satisfactory explanation. Hence, to advance the minimum wage research program this article synthesizes and formalizes from Lester's writings an alternative institutionalist model of labor markets, represented in a set of five diagrams. The model provides considerable new insight and explanation for a zero/near‐zero employment effect, anchored on lack of a well‐defined labor demand curve, alternative cost‐absorbing managerial actions, and aggregate demand effect.

Keywords::

• minimum wage
• institutional economics
• labor market
• employment relationship

JEL Classification Codes::

• B25
• J20
• J23
• J42
• J88

Notes

1 In the book, Card and Krueger (1995) present their empirical results, interpret the implications, but never take a for/against policy position. Nonetheless, the conservative/libertarian reaction was sometimes vitriolic. Illustratively, Nobel‐laureate James Buchanan (1996) wrote of Card and Krueger, “Fortunately, only a handful of economists are willing to throw over the teaching of two centuries [law of demand]; we have not yet turned into a bevy of camp‐following whores.” In the same vein, Walter Block (in Cappelli and Block 2012, 15) declares Card and Krueger's study, “has been subject to such withering and devastating rebuke that it's a wonder that these guys didn't go the way of Michael Bellesiles [academic historian found guilty of creating fake data] in losing tenure.”

2 Since U.S. labor economics from the 1890s to 1960s was dominated by the IIR approach (McNulty 1980), Lester was orthodox in labor economics but heterodox with respect to the microeconomic price‐theory core of the discipline. The focal point of IIR critique was the commodity/auction conception of labor markets and applicability of the demand/supply model (Pierson 1957). The neoclassical economics (NE) school, led by Chicago economists, reclaimed labor economics after the 1970s and reinstated the competitive model as the theoretical foundation. The institutional vs. neoclassical divide in labor economics refracts into the “first” (IIR) and “second” (NE) law and economics movements in the labor/employment law field (compare Wachter 2012 and Kaufman 2012a).

3 In IIR, labor markets and firms are endogenous human‐constructed, resource‐using governance institutions created by people to serve certain interests with human‐determined boundaries and rule sets, while in NE they are analytical constructs (trading areas, production functions) defined in mathematical space with (competitive) labor markets presented as exogenous, nature‐given, zero‐cost non‐institutions. Anti‐minimum wage economists (e.g., Neumark 2002) mistakenly claim NE theory is objective and value‐free which conflates its internal logic (objective) with its assumptions (normative). NE theory has a free‐market/free‐lunch bias because the market institution is assumed to have zero set‐up/operating costs (hence no cost curves) while, asymmetrically, firms (etc.) have these costs/curves and thus are inherently second‐best to markets (Williamson 1985; Kaufman 2013b). The one‐sided logic of the competitive demand/supply model implies (Boeri and van Ours 2008, 26), “The social optimum coincides with the laissez‐faire equilibrium” and (18) “all labor market institutions introduce a wedge between labor demand and supply … and reduce the size of the economic pie.”

4 Lester takes imperfect competition in labor markets as an empirical fact, which precludes firm‐level labor demand curves (and labor demand/supply model) for the reasons given. Coasean institutional logic unknown to Lester theoretically establishes the same result (Kaufman, 2010b, 2013b); that is, perfect competition assumes zero transaction cost and constant returns to scale and, since “buy” everywhere dominates “make” as the efficient production coordinator/integrator, the general equilibrium structure of production vertically dis‐agglomerates to the irreducible level of single‐person firms (i.e., an economy of perfect decentralization, such as family farms and sole‐proprietor stores). Single‐person firms do not have employees and therefore have no demand function for labor (in the factor market sense), and hence the economy has no labor markets. Henry Ford can still build large factories and staff them with thousands of assembly line workers, although hired in intermediate product/service markets as independent contractors.

5 The author has done two Lester‐esque field studies of fast‐food restaurants (Young and Kaufman 1997; Hirsch, Kaufman, and Zelenska 2015). Managers/owners uniformly didn't use marginalist first‐order type of calculation to determine employment level, and smiled bemusedly or looked exasperated when I led them through the logic. As Lester found, they perceive MC is constant/declining and the key to profit is increasing sales volume to exploit declining ATC. A McDonalds did twice the volume of a nearby Burger King and was more than twice as profitable, with some positive feedback into wages/training/benefits. The McDonalds weathered the 2008 financial crisis but three of the Burger King units owned by the franchisee closed. Because friendly customer service and minimal wait times are crucial to repeat business, maintaining morale/team spirit is also crucial which explains reluctance to cut hours/headcount (commodity logic). Instead of W = MRP_L calculation, managers use a labor hours/sales measure, L = αSales (per a Leontief production function), as a guide to employment/hours with an average‐based, mark‐up rule of thumb not to let labor compensation exceed, say, 22%–24% of operating cost.

6 Young and Kaufman (1997) found wage dispersion for hamburger fast‐food restaurants located on opposite corners of a street intersection of about 5% for the entry‐level wage, 10% for the crew wage, and roughly 15% for all sampled restaurants within a given mile radius from the city center. In the regression analysis, sales volume was the strongest predictor of the restaurants’ wage level, along with distance from city center (a positive wage gradient from the center).

7 Human productivity differences are decomposed into two distinct parts in the diagram: the first part comes from behavioral plasticity for a given set of characteristics and is represented as a band, the second comes from differences in cost/productivity characteristics and is represented as separate cost lines.

8 One minimum wage‐induced sales effort of a local fast‐food franchise owner interviewed by the author was to market the stores as a place to hold children's birthday parties and church groups. A NE response is to ask why the owner did not initiate the program earlier if it brings in sales/profit? An answer is the time, energy, and brains of the owner and unit managers are scarce resources and can't discover/implement all available cost‐reduction/sales‐expansion actions. Another answer is the NE assumption that competitive pressure and owner's desire to maximize profit ensures firms are on their minimum cost curves (Stigler 1946) begs the question because it rests on two free‐lunch and impossibility assumptions: cost‐free/self‐enforcing market institutions and unbounded cost‐free information/decision‐making—which contradict the foundational condition of scarcity.

9 Upward bias is also built in by counterfactually assuming the labor market starts out at demand/supply equilibrium. The Coase theorem also asserts in perfect competition the employer/employee can bargain around the minimum wage with a wealth effect but no employment effect. Free‐market logic, symmetrically applied, also implies open immigration.

Additional information

Notes on contributors

Bruce E. Kaufman

Bruce E. Kaufman is a professor of economics at Georgia State University, Atlanta, Georgia.