Why Is Wage Growth Stagnant?: An Analysis of 10 Theories Behind the "Wage Puzzle"

Last week, the Bureau of Labor Statistics announced that the U.S. economy added 157,000 jobs, and that the unemployment rate edged towards 3.9 percent. The U.S. unemployment rate has not been this low since 2000. Not only has the unemployment rate been quite low, but the economy is growing more than anticipated. Just last quarter, the GDP grew 4.1 percent. Granted, the reasons for why are debated, but we are seeing some good macroeconomic indicators. There is one macroeconomic indicator that is perplexing economists and public policy experts: the wage growth. Basic economics tells us that when the demand of a good or service increases, so does the price. In this case, as the demand for workers increases, wages also increase because employers need to attract increasingly scarce labor. This is known as the Law of Demand. And yet, what we have seen since the 1970s is wage stagnation. This has been notable since the economic recovery from the Great Recession (see below). This phenomenon has been labeled "the wage puzzle." Many have tried to figure out what is causing the stagnation, including those at the New York Times and Forbes. What I would like to explore here is whether we could solve the puzzle, or at the very least, see if some pieces of the puzzle fit better than others.

1. More labor market slack: This argument assumes that the unemployment rate is higher than we think. In order to measure official unemployment, we use what is referred to as U-3 unemployment. I have discussed the limits of this official unemployment measurement, but essentially, U-3 unemployment overstates the health of the market because it does not include those who left the labor market. If the Great Recession did indeed drive people out of the labor market but are willing to rejoin the labor market, that would mean that the official unemployment rate paints too rosy of a picture of our economic health. This could explain how the unemployment rate and wage growth can simultaneously be low. However, there are some issues with this theory. One is that we are unsure as to whether prime-age employment (i.e., 25-54 years of age) is below the long-term trend. As a matter of fact, we see there is a good correlation between prime-age employment and wages (see below). The second is that other macroeconomic indicators (e.g., number of job openings, rate at which workers voluntarily quit their jobs) point to a robust labor market. Essentially, the U-6 unemployment rate and the prime-age employment rate are not quite as good as they could be, which means there is some slack. Even so, I do not think that there is so much slack that it is the main piece of the wage puzzle.

2. Substitution Effect - Retirement of Baby Boomers: Evidence from the Federal Reserve Bank of San Francisco implies that the wage stagnation is caused by a structural shift in workers (Daly et al., 2016). As Baby Boomers retire, younger workers are entering the market. Since younger workers have less experience and skills than those retiring, they do not get paid as much as their more senior workers, hence the stagnation.

3. Increase in non-wage benefits: One of the flaws at looking at wage growth is that looking at wages excludes non-wage benefits. The American workplace has experienced an increased prevalence in health benefits, retirement benefits, child care, employee discounts, and vacation time. I think this is convincing when considering healthcare costs. Most employees are on what is known as employer-sponsored health insurance. This tax break drives up healthcare costs in a way no other country experiences while exacerbating income inequality. It is so bad that three years ago, I called it the worst tax break in the American tax code. However, I do have a concern with this explanation. In 1987, the BLS found that benefits account for 26.8 percent of overall worker compensation. By 2018, that figure increased to 31.8 percent (Table 1). Looking at these figures, I think that non-wage benefits are a piece of the puzzle, but hardly the largest one out there.

4. Lower productivity: Basic microeconomic theory postulates that wages are primarily determined by productivity. In the past ten years, productivity growth has been dismal relative to previous years, as the BLS and Vox pointed out. The IMF recently released research going as far as stating that "weak wage growth is mostly a product of low productivity growth (Abdih and Danninger, 2018)." The Right-leaning Heritage Foundation released some research showing how experts across the political aisle have found a strong connection between productivity and wages (Sherk, 2016). A working paper led by a Princeton University professor (Grossman et al., 2017) suggests that productivity slowdown is part of the puzzle because "when human capital is more complementary with physical capital rather than raw labor....[it] can itself lead to a shift in the functional distribution of income away from labor and towards capital." Given that the IT sector involves greater capital intensity, this explanation is certainly feasible (Karabarbounis and Neiman, 2017).

5. Increased Market Concentration and Monopsony Power: A paper from MIT argues that increased market concentration, from companies known as superstar firms, has contributed to the reduction in labor's share of total income (Autor et al., 2017). A report from the National Bureau of Economic Research argues that from 1979 to 2009, markets have become more concentrated (Benmelech et al., 2018), as does a paper from Harvard Law Review (Naidu et al., 2018). With less employers in the market, employers have more power to set wages, hence the monopsony-like power. Part of this trend is due to increased emphasis on inorganic growth (e.g., mergers, acquisitions), but much of it is due to the collusion between Big Business and Big Government.

Granted, this trend of market concentration varies from market to market, but research shows that 23 percent of Americans are in moderately or highly concentrated labor markets (Azar et al., 2018). Also, the research on monopsony power is relatively new, which makes it difficult to determine the extent to which this affects stagnant wages. Nevertheless, given the amount of time I have spent doing market research professionally, I can safely say that my experience backs up the notion of increased market concentration, hence making it a piece of the wage puzzle.

6. Backlog of wage cuts from Great Recession - Wage Rigidity: During recessions, it is typical for employers to cut wages in order to correct for the economic downturn. After the recession is over, wages go back up. However, we did not experience that during the Great Recession. This economic phenomenon is known as downward nominal wage rigidity. It caught the eye of the Federal Reserve Bank of San Francisco to the point where they assumed it to be a major contributing factor post-Great Recession (Daly and Hobijn, 2015). If this theory is indeed true, then we still have employers whose ability to respond to economic conditions (whether good or bad) has been impaired.

7. Corporate tax rates: Another theory, this one coming from the Right-leaning American Enterprise Institute, postulates that wage growth is being driven by corporate tax rates. AEI found that the tax incidence of the corporate tax falls primarily on the workers. By lowering the corporate tax rate, wages will increase because workers will foot less of the bill. Don't get me wrong. I think the corporate tax rate should be lower (see here and here). A permanent tax cut was one of the features of the GOP tax bill I found appealing. Nevertheless, I am skeptical of this being a [major] piece of the wage puzzle for two reasons. One is that both statutory and effective corporate tax rates decreased between the 1970s and now. The second reason is that wage stagnation is not strictly a U.S.-based problem: it affects the global economy, including countries with lower corporate tax rates. As much of a proponent I am of a lower corporate tax rate, I am doubtful as to whether or not it would ameliorate wage stagnation.

8. Increase of women and foreign-born workers in the marketplace: The libertarian Mercatus Center theorizes that the influx of women and the influx of foreign-born workers is a cause of a wage stagnation because less experienced and skilled workers drag down wage growth. I would find it more believable that women entering the workforce would have this effect since it was during the 1970s that women entered the workforce at significantly higher rates. However, more than a generation has passed since then, and women have become more competitive in the workplace. I'm not so sure about foreign-born workers. When looking at H1-B visas, for example, we see that higher-skilled foreign-born labor tends to get paid less. A similar phenomenon is observed with lower-skilled foreign-born workers. I would argue that the influx of foreign-born workers presently has more of an impact on wage stagnation than women entering the workplace.

9. Decreased unionism: One of the favorite arguments of the Left is that the decline of unions has caused many adverse effects, including wage stagnation. Because private-sector union membership declined from 21.2 percent in 1980 to 6.5 percent in 2017, wages have stayed stagnant, or so argues the Left-leaning Economic Policy Institute. The main problem with this argument is that union membership as a percent of the overall workforce was declining more than a decade before wage stagnation became an issue in the 1970s (Dinlersoz and Greenwood, 2012). Even the IMF thought it was a relatively small contributor, especially in comparison to automation and technological development (Abdih and Danninger, 2017, p. 16). Speaking of which.......

10. Automation and technological development: In the aforementioned 2017 IMF study, the authors found that automation was the single largest contributing factor to wage stagnation (p. 21). The theory behind this is that technology has allowed for increased automation. Technological development, particularly in information and communication technology, made capital easier. As such, there was a decline in the labor share (p. 6). Automation had such an effect that it was the single largest factor in contributing to the decline of employment in the U.S. manufacturing sector.

Conclusion
There were some other theories out there, including the gig economy, globalization, increased usage of noncompete clauses, and minimum wage laws. Nevertheless, I wanted to cover the theories that were most prevalent. Even with these theories, I have three reasons to be skeptical about excessively worrying about the wage puzzle:

1. Moody's economist Adam Ozinek found that when looking at the employment rate (as opposed to the unemployment rate), wage growth is where it ought to be.
2. When adjusting for inflation and including the amount that employers contribute to benefits, there is not wage stagnation. That is not my conclusion, but rather that of a scholar at the Urban Institute, an organization that tends to lean towards the Left.
3. Looking at wage growth in isolation does not take into account purchasing power or the fact that consumption has increased both in quantity and quality over the past 40 years (Sacerdote, 2017). Without looking at consumption or purchasing power, wage growth is divorced from context.

Assuming that we, as a society, decide to label wage stagnation as a problem, how to resolve it would depend on which pieces of the wage puzzle you define as the biggest. For me, the arguments that are most convincing are low productivity, retirement of Baby Boomers, technological development, and increased market concentration. Even if we could agree on which pieces are largest, there would still be a matter of policy prescription. Naturally, I believe that policy should help as many people as possible to live their lives as fully as possible. On the other hand, I am unsure how these issues will be addressed. As progress is further made, I hope to discuss the matter in more concrete terms. In the meantime, I hope this provides some insight to a complex, interconnected topic.