Note: This post is not about whether there should or should not be a minimum wage or, if there should, whether the federal government should set one. This is a quick examination of some facts.
The U.S. federal minimum wage has been $7.25/hr since July 2009. Working 40 hr/wk, 50 wk/hr results in an annual pre-tax income of $14,500. Working 50 hour weeks every week provides $18,850.
(You can skip the description of the metrics and jump to the bottom if you already understand the pros and cons of various normalizing methods.)
There are several ways to think about about today’s minimum wage vs. historical values.
- Adjusted for inflation. This view adjusts to a constant standard of living, only accounting for the loss of purchasing power of a dollar over time. In other words, if today’s value were the same as 50 years ago, then the minimum wage worker today would have the about the same standard of living as one 50 years ago.
- Adjusted for productivity growth. This view shows what a previous value in the minimum wage would be today if minimum wage workers enjoyed wage increases equal to inflation plus the economy-wide increase in worker productivity. This view shows the wages necessary to increase the standard of living of minimum wage workers at about the same rate as economic growth.
- Compared to the median wage. We can compare the annual earnings of a minimum wage worker to the median worker’s annual pay. Since the median pay is the point at which 50% of workers earn more and 50% earn less, it provides a reference vs. the “middle” of the income distribution.
- Compared to the average wage. This is similar to the median wage comparison, but instead of using the income distribution, it compares against the average wage that includes all wage income from janitors to doctors and investment bankers.
Pros and Cons of these views:
- Adjusted for inflation. This is easy to calculate against data with a long historical record (inflation). It is a good measure of the absolute changes in the standard of living a minimum wage job would afford. The downside is that it implies to the casual reader that the minimum wage should freeze standards of living. In other words, a flat line may appear to indicate that the minimum wage is unchanged, but in fact it would indicate that minimum wage workers are falling further and further behind other workers.
- Adjusted for productivity growth. This has the advantage of showing what minimum wages would be like if all workers shared in the benefits of higher productivity. Unfortunately, it necessarily relies on a given start date and implies that both the minimum wage and the income split between capital and labor at that initial time is \ appropriate. The curve may change substantially based on the choice of start date. Also, this method combines issues that effect workers broadly (income distribution and labor-capital split) and embeds them in the presentation of the minimum wage.
- Compared to the median wage. The median wage is probably a good reference for the minimum wage worker, since the median wage may be a good indicator of a worker’s aspiration and social comparison. By accounting for changes in the income distribution over time it avoids including the large concentration in income going to the Top 10% of wage earners. Median wage data are reliable, but may not fully capture all forms of compensation and the middle.
- Compared to the average wage. Like the median wage approach, this has the advantage of using the economy’s actual split of income between capital and labor each year instead of implicitly fixing it like the productivity method. However, this method is not influenced by changes in distribution of wage income. So we can better see how a minimum wage worker’s income has changed vs wages in the economy as a whole.
Note: All of these methods are properly per-capita.
OK, here are some links, charts and comments
Inflation Adjusted View. The linked chart shows the real federal minimum wage since it was first enacted in 1938. We can see that it was basically flat during WWII and the inflationary 5-year period following the war. Then the real minimum wage rose steadily for 20 years. After peaking around 1970 it declined as increases failed to match high inflation during the 1970s and early 1980s. It has been flat for the past 25+ years and today is at the same level as the mid-1950s. (Shadowstats conspiracy theorists would calculate a huge collapse in the real minimum wage that would put today’s levels much lower.)
Productivity Adjusted View. Using 1947 as the base year the linked chart and article shows that the minimum wage would currently be $17/hr if minimum wage workers shared received gains from productivity growth. This is 135% higher than the current value of $7.25/hr.
Median Income Comparison View. The graph below (click to enlarge) compares annual minimum wage earnings with median household income since 1984. I was unable to find reliable median per worker pay data and the household series was only available back to 1984. The relationship today is about where it was 30 years ago, though there was some volatility within a band.
Average Income Comparison View. The graph below (click to enlarge) compares annual minimum wage earnings with median household income since 1951. The relationship was in a band centered on 50% for about three decades and then rapidly and steadily declined through the 1980s and 1990s before leveling out around 30%. So compared to the average worker’s annual compensation, minimum wage workers have seen a large drop in income. (The average wage data come from the Social Security Administration’s AWI data and represent average per worker compensation.
Considering the charts together
By looking at the charts against one another we can see a couple of things. First, the widening gap between minimum wage pay increases and productivity increases is not an issue specific to the minimum wage per se, but rather captures a) more income flowing to capital and b) more wage income concentrated at the upper income levels. Both changes are required to explain the large gap in the productivity-based view vs. the average wage view. Comparing the median and average wage view shows that the decline in minimum wage pay vs. overall wages is really about the broader issue of income distribution rather than the minimum wage taken in isolation.
Comparison with the Obama proposal of $10.10/hr:
Inflation-adjusted view: The proposed rate would be very close (about 5% less) to the 1968 peak.
Productivity-adjusted view: The proposed rate would be more than 40% below the productivity-adjusted figure.
Median-income view: The proposed rate would increase the minimum wage level relative to median income above any level seen in the past 30 years. In fact the minimum wage income would be about 30% higher than it’s been relative to median income.
Average-income view: The effect of the proposed rate vs. average wages would be similar in magnitude to the median wage impact. However, the data series allows us to see that the level would remain below the level in the 1950s through the 1970s.
I will leave it to the reader to consider the knock-on impacts of leaving the minimum wage at its current level (ongoing real decreases), increasing with inflation, or increasing aggressively to a real level and income share not seen in decades.