Monday, July 3, 2017

Seattle and Denmark: More Proof of Minimum Wage's Harm to Low-Wage Workers

The minimum wage battle continues. In June, three newsworthy studies on the minimum wage were released, which is quite a bit in such a short time. One of the reports was on minimum wage in Denmark, and the other two were on the effects of minimum wage in Seattle. We'll get back to the Denmark report, but the Seattle reports merit background information. In 2014, the City of Seattle was the first city to pass a bill to raise the minimum wage to $15/hour. Beforehand, the minimum wage was $9.47/hr. It was raised to $11/hr in 2015, to $13/hr in 2016, and reached $15/hour as of January 1, 2017.

To show the effects of Seattle's minimum wage hike, its City Council hired economists from the University of Washington (UW) to conduct the study (Jardim et al., 2017). It's not as if University of Washington is some second-rate school. US News ranks its graduate-level economics program 35th in the country, which is not terrible by any means. It's a reputable institution, but yet is was still a problem, a problem which had nothing to do with the institution's reputation.

City Council member Kshama Sawant, who is a Socialist politician, did not like the results of the initial draft of the study. Sawant claimed the research methodology was flawed, although renowned economist David Autor found the UW study to be "very credible." The last thing a politician needs is a failed newsworthy policy months before municipal elections. Ideology and election season help explain why Sawant decided to contact Michael Reich, an anti-capitalist professor at the University of Berkeley. Fox News obtained emails between Reich and the City Council to schedule the release of the Berkeley paper (Reich et al., 2017) before the UW paper in attempts to discredit the UW paper.   (Side Note: The Albany-Union Times also found that the Berkeley research team has coordinated with minimum wage advocates to create supportive reports, so it's not just a Fox News conspiracy).

Yes, it is true that UW had some conflicts with the City Council beforehand, but this is still an example of selection bias. The City Council did not like the results they wanted to hear, so they sought out a Left-leaning, anti-capitalist economist at the University of Berkeley to get the desired results. What was so upsetting about the UW minimum wage findings? And how does the Denmark minimum wage report relate to the Seattle incident?

Before delving into the studies and their findings, I want to add this caveat: Given the nature of minimum wage, there are going to be more methodological issues than other public policy studies. Why? Because economists are trying to figure out how the minimum wage interacts with the greater economy. Much like I explained with Kansas' tax cuts last month, there are multiple factors in play, including taxes, regulations, economic performance of surrounding areas, industrial composition of a given city, the list goes on. As we'll see shortly, we'll find shortcomings for these studies. That does not, however, mean we cannot draw meaningful conclusions from the findings.

University of Washington Findings
The UW study was not flattering because it had a number of findings that were flattering for minimum wage proponents, including:
  • The second wage increase to $13 reduced hours worked in low-wage jobs by 9 percent.
  • Although the hourly wage increased by 3 percent, overall wages decreased by $125 per month for low-wage workers, which would mean a $1,500 per annum pay cut per person, or a $120 million loss for the City of Seattle (not huge, but also still worth noting with a city that has a $231 billion economy).
  • A loss of 3.5 million hours worked per calendar quarter, which is an annual total of 14 million hours lost. 
  • Low-wage jobs declined by 6.8 percent, which means a loss of 5,000 jobs. 
  • Low-wage labor demand has an elasticity of -3.0 (p. 35), which is the economic way of saying "low-wage labor is easily replaceable."
If I were for an advocate for minimum wage, I would want to discredit this study to no avail, much like the Left-leaning Center for American Progress (CAP) and Economic Policy Institute (EPI) attempted (see criticism of EPI's criticism here), which is even more hilarious considering that one of the economists on the UW research team is a former EPI employee. The most valid and prominent complaint the Left-leaning think tanks had was that it excluded many multi-site firms, which removes 48 percent of low-wage earners. At first glance, this comes off as a huge misstep on UW's part. Fortunately for UW, they covered the reason for this omission (p. 14). Not only that, the omission does not negate the findings, or at least the magnitude of the findings, since multi-site firms were more likely to plan and implement staff reduction as a result of the minimum wage (p. 15). The only way that way the exclusion of multi-site firms would negate the results is if they massively expanded operations, which again, is implausible given the reported staff reductions. CAP also brings up that Seattle's tech boom is causing the increase of its higher-wage jobs, which means the adverse results of the minimum wage hike are overstated. This would be backwards because if Seattle's economy is growing faster than expected, imagine what we we would see if the economy were not booming, much like during the Great Recession.

I do, however, find merit with CAP's critique that the UW study is just one study that covers one city. It does not automatically denounce minimum wage increases. It could simply apply to Seattle in general or Seattle in this specific time period. I would say there is enough academic literature to support the idea that the UW study findings can be broadly applicable, but I think there is an even more important point that gives the UW study more weight than past minimum wage studies. This study was unique in that it was able to do something past studies were not able to do: permit the direct observation of hourly wages (Jardim et al., p. 34). Previous studies used proxies to get at that data on hours and wages. Thus, the study allows for more direct and precise measurement of the effects of minimum wage that past studies did not have.

While it is just one study on one city that does come with methodological flaws (because honestly, what study doesn't have methodological flaws?), it still carries with it a good amount of credence in terms of showing the effects that minimum wage hikes can have on an economy because it is conducted by an apolitical group and is done so with superior data.

UC-Berkeley Study on Seattle
The UC-Berkeley study shows the opposite results of the UW study, which is that minimum wage hikes have no real effect on employment or hours worked for low-wage workers. I am already skeptical of the UC-Berkeley study based on the politicizing that went into making the study a possibility. The study suffers from more flaws than the political, the major one being that it only looks at the restaurant industry. It is hypocritical for Reich to criticize the UW study for omitting 48 percent of low-wage workers while Reich himself omits 70 percent of low-wage workers in his own study by analyzing restaurants only.

But the Left could argue that looking at the restaurant industry has been standard in conducting minimum wage research. That standard came from the Card-Kruger study in 1994, which has been since scrutinized (see here, here, and here). Interestingly enough, when the UW economists looked at the restaurant industry by itself, it came to similar conclusions that Card-Kruger came to in 1994 and that Reich came to in his Seattle study, which is that there are little net effects in the restaurant industry (Jardim et al., 2017, p. 35).

The UC-Berkeley study is inferior for two reasons besides the political contentiousness surrounding its creation. One is that Berkeley uses data from the food industry only, which represents a fraction of low-wage workers in Seattle. Much of past minimum wage literature suffers from only looking at one industry or looking at one demographic (e.g., teenagers), which makes it tenuous. The second, as previously mentioned, is that the UW-research team had detailed work hours data that the UC-Berkeley team did not have. The Berkeley study is nothing more than a political ploy to allow Seattle politicians give themselves a pat on the back before their re-elections.

Denmark Minimum Wage Study
In early June, The Centre for Economic Policy Research released a study (Kreiner et al., 2017) on Denmark's minimum wage, specifically with regards to youth minimum wage. Denmark has a peculiar minimum wage law in which the minimum wage increases for a Dane by 40 percent when they reach their eighteenth birthday. What happens in this case? Unemployment for this age demographic drops, and it takes two years before the age-specific employment rate is recovered (see below). What is even eerier about these findings is that it takes college and apprenticeships into account (p. 18), which is to say the drop in employment is pretty much due to the minimum wage hike (p. 3). Much like with the UW study, CEPR had similarly granular data to come to its conclusions. This study is also significant because it shows a) the negative effects of a sudden increase in minimum wage, and b) the minimum wage disproportionately affects youth [because they have less skills and experience] (p. 1).

The Denmark study might focus on youth employment, but the connection between the Seattle and Denmark cases studies is the following. When you raise the minimum wage modestly, you'll have modest impact on the economy. When you raise the minimum wage more drastically, the effects will be more drastic. For those who are fiscally conservative, this comes as no shock. As conventional economics teaches, when the cost of a good, input, or service is increased (e.g., carbon tax), the consumer (or employer) wants to consume less. The same goes with the minimum wage: when you increase the cost of labor, the employer hires less workers (or alternatively, increases product price on consumers, reduces workers' benefits, or reduces hours worked).

What is worrisome about the Seattle study is that the UW researchers did not even measure the impact of the full hike yet because the hike to $15 only took place this past January, so we do not even know the extent of the economic damage the hike will cause. By raising the minimum wage from $9.47 to $13 (or 37 percent) in a matter of a couple of years, we already see the economic impact. It is no surprise that a significant increase in minimum wage significantly hurts the very people minimum wage laws were meant to help. While neither Seattle study has been peer-reviewed, it is clear that the results will have implications for the minimum wage debate across the country. If I had to take an educated guess, things just got a lot more difficult for minimum wage advocates, especially in light of the more detailed wage and hour data that was available in the UW study that has not been available in past minimum wage research. At the very least, this should put the kibosh on the "Fight for $15" advocates, and at the most, this should be the beginning of minimum wage advocates to re-examine the magnitude of the negative impacts that minimum wage laws generate.

No comments:

Post a Comment