For those following me on Twitter, you might know that I’m a bit fixated with inflation, its underlying causes, and how best to arrest it.
Here’s what we know after one year’s worth of controversy: (1) inflation is largely coming from two sectors of the economy, housing and electricity; (2) inflation does not seem to be slowing from interest rate increases, with core inflation of 6.6 percent hitting a 40-year high in September; and (3) neoliberal economists cannot fathom a theory of inflation that doesn’t blame workers.
I will be writing plenty on alternative remedies to arresting inflation, including de-concentration efforts, temporary industry-specific price controls, and FTC enforcement of invitations to collude. But in this piece, I wanted to stick a fork into the traditional view that inflation is best tempered with interest rate hikes.
The theory behind this view is that the labor market serves as a barometer to how hot the economy is running. As unemployment declines, the story goes, workers demand higher wages due to their newfound bargaining power. Profit-maximizing firms raise the price of their products in response to rising labor costs, yielding an inverse relationship between unemployment and inflation, as predicted by the Phillips Curve. Higher interest rates make investment projects less attractive by raising the cost of capital. Slow the labor market by making investments less attractive, demand for workers needed to staff new projects will decline, wages will fall, demand for goods and services by those underemployed workers will decline, and the rest of the economy will naturally cool down.
This theory is badly antiquated. There may have been a time when output and labor markets were competitive, prices were set at marginal costs, and wages were set at a worker’s marginal revenue product. But that time, if it ever existed, has come and gone.
In many industries, particularly concentrated ones with high fixed costs—or industries dominated by what some economists have dubbed “superstar” firms—prices are untethered to labor costs. A pharmaceutical company doesn’t set its drug price based on the salaries of researchers. A gas station doesn’t set its price based on the attendant’s hourly wage. A hotel doesn’t set its nightly rate based on the wages of the cleaning crew. A car rental company doesn’t set its price on the wages of the worker sitting near the checkout booth. A university doesn’t set tuition based on the wages of its adjunct professors. I could go on.
Given the fixed-cost nature of these industries, many prices in the economy are set to maximize revenue, which is purely a function of the demand elasticity facing the firm. Wages don’t enter the pricing calculus, as they too are perceived (at least in part) as a fixed cost.
Now I will grant that in certain labor-intensive industries, such as coffee shops and diners, prices will be tethered to labor costs. But these industries are not the drivers of inflation. Inflation is coming from concentrated industries such as food production, energy, and rent (more on that in my future work).
Even in some labor-intensive industries, such as sports, the price for pay-per-view does not rise with player payouts. Prices again are set based on consumers’ willingness to pay for the product. The players and owners divide a fixed pie, and an increase in the players’ share does not cause prices to rise.
Another reason why wages and prices have become untethered, aside from the evolution of our economy to high fixed cost industries, is growing monopsony power and the accompanying erosion of worker bargaining power. Employers with buying power drive a wedge between their workers’ marginal revenue product and wages, based on the (low) elasticity of supply they face. This markdown below marginal revenue product is the same phenomenon as we see when monopolists mark up prices above marginal costs in inverse proportion to the elasticity of demand they face.
When productivity increases, an automatic increase in wages no longer follows, as monopsony power now disturbs the the mapping from productivity into wages. Put differently, monopsony power has weakened, if not entirely severed, the linkage between prices and wages. Employers with buying power don’t need to acquiesce to wage demands: companies like Amazon and Starbucks have fought desperately against offering even the most meager concessions to workers. And even if they do, they already enjoy a sufficiently comfortable profit cushion that they don’t need to automatically revisit prices. Hence, an increase in wages won’t necessarily lead to an increase in prices.
But don’t take my word for it. A recent paper by two economists, David Ratner and Jae Sim of the Federal Reserve Board, shows that labor market policies that have eroded worker bargaining power are the source of the demise of the Phillips curve, which predicts that inflation rises as unemployment falls. As unemployment fell from 6.5 percent in 2009 to 3.5 percent in 2019, however, inflation showed no signs of increasing, or what is called “the missing inflation” puzzle. Ratner and Sim conclude that the “collapse of workers’ bargaining power” flattened the Phillips curve. The authors point out that, using their models in which workers bargain to keep the price-cost markup as low as possible so that wages and the labor share would be larger, the decline in worker bargaining power can explain the secular decline of the labor share, the secular rise of the profit share observed since the 1980s, and the stock market capitalization ratio.
Take the case of Starbucks, which is fighting off an attempt at unionization. Given the lack of bargaining power among its workers, when Starbucks raises the price of a latte, most of the price hike falls to the bottom line. If the workers were to unionize and demand a greater share of the pie, Starbucks might be less inclined to raise its prices, knowing that some of those gains would now be shared with workers. Continuing this union hypothetical, if workers were to become scare due to a decrease in unemployment, the union could extract higher wages, which likely would be passed on to consumers in the form of higher latte prices—revealing a steeper Phillips curve.
A flattened Phillips curve means that a decrease in unemployment that should, in theory, give workers greater bargaining power, doesn’t have any affect on prices (even if wages modestly rise) and thus inflation. Why? Because the power imbalance is so skewed in favor of employers that the small increase in the scarcity of workers doesn’t materially move the price needle (let alone the wage needle). In the post-Covid era, wage inflation peaked in March 2022 at just 6.8 percent, but for the reasons explained above, that change likely didn’t contribute to inflation, which outpaced wage growth in the same month. As explained by ESI’s Josh Bivens, “If the only change in the economy over the past year had been the acceleration of nominal wage growth relative to the recent past, then inflation would be roughly 2.5–4.5% today, instead of the 8.6% pace it ran through March.”
A flattened Phillips curve also implies severe labor market dislocations would be required to arrest inflation via rate hikes. As explained by Seccareccia and Romero in an INET blog, “the flatness of the relation [between unemployment and inflation] implies an immensely high sacrifice ratio if pursuing a Volcker-style strategy … since it would require excessively high unemployment to get the inflation rate down by even a very small increment.” Former Treasury Secretary Lawrence Summers claims that we now need five years of six percent unemployment to curb inflation.
Despite these horrific tradeoffs, the church of neoliberalism led by Summers, preaches that monetary policy that reduces worker power (by creating unemployment) will eventually albeit painfully lead to lower prices and slowing inflation. Ironically, Summers co-authored a Brookings research paper in 2000, before inflation took hold, in which he argued that to arrest the decline in labor share, “institutional changes that enhance workers’ countervailing power—such as strengthening labor unions or promoting corporate governance arrangements that increase worker power—may be necessary (but would need to be carefully considered in light of the possible risks of increasing unemployment).” It is not clear whether Summers would continue to advocate for worker power in an inflationary environment.
In sum, managing inflation via interest rate hikes is an outmoded school of thought that may have worked when labor markets were more competitive and the power balance between workers and employers was evenly split. Given the severing of the nexus between wages and prices, it is time that we abandon the old tools and explore alternatives that arrest the source of price hikes at their source. This will be the subject of future pieces.