January 22nd, 2021

Inflation, Specific and General

—   Andrew Elrod

Inflation, Specific and General

—   Andrew Elrod


Liubov Popova, Textile Design, 1924.


Until 1980, the annual rate of change of the Consumer Price Index (CPI), the weighted measure of the cost of a basket of core consumer goods, increased at an accelerating pace in every business-cycle expansion, reaching double digits during the 1940s and 1970s. Inflation—its causes and consequences—was at the heart of economic debates throughout this period, when the discipline of macroeconomics took its current form. While we understand individual industry price changes in terms of supply, demand, and market power, our conceptual tools for understanding inflation remain weak. Neither monetarist attention to the “money supply” nor the institutionalist focus on the Phillips Curve have provided a reliable guide with which to construct macroeconomic policy: the relationship between the price level and the unemployment rate has attenuated with the decline of collective bargaining, and a more than eleven-fold increase in the sum of checking deposits, savings deposits, and cash on hand in the US since 1980 (the standard M2 measure of the money supply) has seen the annual increase in the CPI fall below four percent for all but a handful of the past forty years. Even the phenomenon implied by the term is the subject of confusion: Does it refer to the amount of money or debt in circulation, or to a rise in price and values? If the latter, which prices or values?

In the decade-long recovery from the 2008 financial crash, Federal Reserve and Treasury officials began to reconsider their models of the inflationary process, as well as the priority of price stability among macroeconomic objectives. The coronavirus pandemic has accelerated both trends. Synthesizing the lessons of the Obama-Trump era, Federal Reserve Chairman Powell noted that “You can have 3 or 3.5 percent unemployment for a couple of years, and really, you see modest moves in wages and almost invisible moves in inflation.” His comment reflects the emergence of an entirely new understanding of the structure of the economy, with potentially significant policy consequences. Reflecting concern over the place of price stability in the public scale of values, the Federal Reserve announced in August of 2020 that it would keep interest rates near zero until the annual rate of increase in the CPI is “moderately above two percent for some time,” while a growing number of international organizations and now senior leaders of the Democratic Party, victorious in the executive and with a slim majority in the legislature, are urging a historic step up in the level of government spending.

Understanding the causes and consequences of inflation is crucial to grasping patterns of distribution and investment. Investors evaluate their portfolio decisions on the basis of the expected future value of returns, a task only possible in relation to future prices. Deflationary expectations tend to postpone investment decisions in physical capital, while rising prices accelerate them. Historically, capital income has increased much faster than wage income and capital’s share of total income rises during the first half of business expansions in the US, as profits swell from a greater volume of sales much faster than employers raise wages.

Managers and owners are unwilling to reduce money profits when workers do succeed in raising wages; inflation is the mechanism for redistributing real income when money claims exceed capacity. Anti-inflationary policies thus have dramatic consequences for workers. The customary solution is to raise the unemployment rate until workers are weak enough to moderate wage demands. But when raising employment is a top priority, that solution moves off the table.

The signals of auspicious policy shifts therefore have some investors concerned. Warn the chief forecasters at Blackrock, the world’s largest asset manager: “Domestic workers could gain greater bargaining power on wages... especially in places where the political pendulum is swinging toward addressing inequality. This may take the form of higher minimum wages and taxes. That would make wages more sensitive to domestic slack and pricing pressures.” Already, logistics prices are rising as Amazon, FedEx, and UPS have raised their margins in a consumer economy increasingly organized around home delivery and weak labor standards. Even older industries such as iron and steel are seeing prices rise, with the spot price of coiled sheet steel more than doubling since August lows.

For the majority of people, for whom wages are income rather than a cost of production, devising alternative anti-inflationary policies is critical. But the attention of macroeconomic theory to the general price level has prevented this thinking from taking place. While the low price of commodities, driven by the massive expansion of global manufacturing capacity, has kept the CPI down since the 1970s, the price of services, which make up two-thirds of the cost of living for US workers, have been rising on a secular trend. Price rises that previously signaled struggles over real income or excessive demand have not passed beyond four percent per year since the 1980s, but the collapse of manufacturing prices has been replaced with a rising price of services. These price increases are structural and, if they accelerate, should not be met with fiscal-monetary contraction.

If governments take seriously their professed commitment to sustaining full employment by expanding fiscal policy in the post-pandemic recovery, wages at the bottom of the labor market are bound to rise. However, given that the cost of services has been rising independently of average hourly wage trends, rising wages at the bottom should not be a reason to slow the recovery. When wage increases do begin to shift production costs, the response will need to move beyond the monetarist and institutionalist approaches to inflation developed in the 1960s. These theories of inflation, which stress the likelihood of a “wage-price spiral,” and the importance of the “natural rate of unemployment” or the “non-accelerating inflation rate of unemployment” (NAIRU), were based on the political economies of the 1950-70s, in which domestic experiments in industrial planning and income policies—and the price controls and wage guidelines they entailed—were chucked aside as impossible projects in social engineering. But today such planning is an urgent necessity. Standing in the way are fundamental misinterpretations of how the last inflationary crisis was finally overcome.

NAIRU & the wage-price spiral in an era of globalization

The theory of inflation advanced during the 1960s and 70s held that the general price level has a vertically asymptotic relationship to the level of employment: as unemployment falls below some level (which can only be determined empirically after the fact) workers will demand higher wages, employers will compensate with higher prices, and investors and savers will respond by requiring higher interest rates or profits. Once this spiral begins, a greater volume of spending will be required to maintain the existing level of employment and output, causing price increases to accelerate. Steady inflation will break into a gallop, unraveling long term business planning in the face of higher price expectations, and producing arbitrary windfalls or bankruptcies until government authorities raise the unemployment rate back to its natural level. Because it is consumer spending which determines private demand for labor, the entire mechanism is set into motion by the government’s manipulation of the level of demand.

This model is an artifact of a world in which a much larger share of manufacturing supply chains were located within single countries and in which services were a much smaller share of GDP. In the US economy circa 1950, rising prices had a direct and observable relationship to wage bargaining, with industry-wide unions representing hundreds of thousands of workers across the entire nation—between 80 and 100 percent of workers in transportation, coal, metal mining, steel, automobiles, meat packing, and rubber; over two-thirds of all production workers in manufacturing.1 Bottlenecks could be found where price pressures persisted through multiple business cycles; breaking bottlenecks to overcome rising prices depended on entrepreneurs entering the market or government planning to restructure industries. Economic policy advice depended on ones’ evaluation of a given market, on the relative costs and benefits of restructuring an industry by entrepreneurs as opposed to public regulation or state enterprise, and on the political coalitions that could be found within Congress for each alternative.

This no longer describes the world today. Globalization has restructured product markets so that domestic wage costs comprise only a small portion of manufacturing costs. Global commodity prices increasingly depend on wages in Cambodia, Thailand or Mexico—or on the geopolitical threat of US military intervention, as Saddam Hussein and Ali Khamenei learned. Whereas the domestic manufacturing industry of old protected profit margins through oligopoly pricing, today the explosion of global manufacturing capacity keeps these prices highly competitive. Profit centers are legally removed from many employers who increasingly invest in asset speculation rather than operations, with a long-term depressive effect on labor demand. In the disorganized and highly unemployed society that followed the collapse of the manufacturing economy, wage structures are finely gradated by employer-determined categories. Increases in one part of the structure may not increase the entire wage bill.

Stagflation of the 1970s

The interpretation of the stagflation of the seventies that has survived this transformation is startlingly ahistorical. As the major OECD study on stagflation (written by Nixon CEA chairman Paul McCracken) reported in 1977, “the immediate causes of the severe problems” of the seventies could “largely be understood in terms of conventional economic analysis [and] avoidable errors in economic policy.” Or as Larry Summers writes, “The first attempts to contain inflation [during the seventies] were too timid to be effective, and success was achieved only with highly determined policy. A crucial step was the abandonment of the idea that the problem was structural in nature rather than driven by macroeconomic policy.”

The real cause for the inflation was structural—as political as it was economic. Rising prices are an index of social tension: whether between consumers for a given quantity of goods or services, between the working class and employers over the income of production, or between nations over the terms of trade. In each case, rising prices reflect rival income claims between groups. In the 1960s, the economics profession understood this. The Kennedy and Johnson administrations’ full-employment program entailed a formal bargain of wage restraint from unionized workers in exchange for sustained full employment and an expansion of the social safety net: Secretary of Labor (and former chief counsel of the CIO) Arthur Goldberg, together with the CEA, secured a national wage policy from the AFL-CIO and won from Congress an increase in the minimum wage and the Medicaid program. It was Lyndon Johnson’s failure to move toward a controlled economy that set off the wage-price spiral; after five years of rising corporate earnings, organized labor demanded price and profit controls in exchange for continued wage restraint.

The trend accelerated under Richard Nixon with the 1973 OPEC oil-price increases. The alternatives to the oil and gasoline rationing that followed was, at most, nationalization of the industry, at least, administration of the controls by someone sympathetic to their success—which Treasury Secretaries George Shultz and William Simon were not. A public option to force down oil corporation profits, proposed by Senators from Oklahoma’s Fred Harris to Massachusetts’s Ted Kennedy, represented a middle path. Instead, the Ford and Carter administrations hobbled through six years of rising oil prices, urging wage and price restraint on industry, and attributing their failure to the “excess demand” of the government budget. In his ethnography of working-class Brooklyn during the seventies, sociologist Jonathan Rieder recorded how obvious the alternatives to stagflation were to the public. “I'd go for public ownership of the oil companies,” an Italian-American Nixon loyalist told him after Watergate, “if I didn't think the national politicians were a bunch of thieves.”

The politics of inflation today

The experience of the sixties brought many US liberals—those favorable to the expansion of the welfare state, usually members of the Democratic Party—to embrace the ideas of conservative economists like Friedman. But to fret about the risks of rising labor costs entailed by fiscal expansion today is to misunderstand or deny the wage-price dynamics of the service economy and the role the state and its budget must play in restructuring core service industries. Rent (or “owners equivalent”) alone constitutes over thirty percent of the CPI, utilities and transportation comprise 9 percent, medical care another 9 percent, food away from home 6 percent, and education 3 percent. Whether private services sustained by federal subsidy, or state-level public programs that have turned to tuition and fees with the retrenchment of state funding, the industries that comprise the core of the US services sector have converged in the past four decades on a sort of decelerated Baumolianism. Financed by some combination of philanthropy, public subsidy, subscriptions, and user fees, operations costs in these industries represent the ballooning pressure of workers who must compete with elaborate managerial bureaucracies and vestigial private financing over their share of industry revenues.

Within the service sector, prices for medicine and education have grown faster than average hourly wages in the entire nonfarm business sector in every recovery since 1975. The result is that in 2019, average health-insurance premiums for employer-based group policies with family coverage were around \$20,000 per year; average college tuition in the US for public institutions granting four-year degrees was \$11,800 in the 2015-6 school year, the most recent available from the National Center for Education Statistics (total annual cost of public higher education—including room and board—was \$26,900). These service prices represent tenfold increases since 1975, when the average hourly wage increased less than half as much. In the most recent recovery, rent joined the group of service prices pushing up the total cost of living, outpacing the rise in hourly wages from 2015 to 2020. These rising prices are structural in nature, and if they are pushed up further by rising wages fiscal-monetary restraint will do nothing to lower them. Full-employment should not be sacrificed in an attempt to lower prices. Instead, anti-inflation policy must pass out of the hands of general employment policy and into specific programs for problem sectors that, as the situation requires, restore tax financing, eliminate monopoly profits, and level cost structures.

During the final years of the Obama-Trump boom, the wages that rose quickest were at the bottom of the labor market. Between 2015 and 2020, wage growth in the first quartile—those earning around \$600 per week, or \$15 an hour for a 40-hour week, in 2019—outpaced the other three-quarters of the labor market. This layer of the workforce is largely below the bulk of workers in the high-cost service industries today burdening the workers’ budget. Median weekly earnings in education were about \$1000 in 2019; in healthcare they were around \$800. Expanding the labor market to bring warehouse and food service workers up to the level of teachers and nurses may raise some prices, but they will not be those prices weighing most heavily on workers’ budgets.

What if higher employment does result in wage increases above the bottom of the service sector? Until public control of these industries expands from its current partial financing toward broad restructuring—to ensure existing revenues in education and healthcare are distributed downward toward the provision of services, for example, and away from administrators and stockholders—the relationship between these industries’ prices and wages in general will continue to be unreasonable for most workers. The alternative to public control in these industries is already at work, whether in the downward wage pressure of “Uber”-izing more professions or in the spread of telehealth and MOOCs in healthcare and education. Such technologies allow employers to expand access to their services and lower prices with less labor—the question is whether this is how we want to manage our core social institutions. But so long as we are concerned about generalized inflation caused by rising wages, such questions will be off the table.

  1. J.B.S. Hardman, "The State of the Movement," in J.B.S. Hardman and Maurice F. Neufeld, eds., The House of Labor (Prentice-Hall, 1951), p. 63. Cf. Sumner H. Slichter, The American Economy: Its Problems and Prospects (Knopf, 1948), who calculated that “Two thirds of the workers in manufacturing, four fifths in construction, four fifths in transportation, and four fifths in mining are organized,” p. 34. 

Title Inflation, Specific and General
Authors Andrew Elrod
Date 2021-01-22
Collection Analysis
Filed Under
In Series

Sign up for the JFI Sources, a weekly digest of compelling research across the social sciences.