Labor market trends and their political implications

Politicians often paint an overly rosy picture of economic data when they are in the administration, and put a negative spin when they are in the opposition. Statistics regarding the labor market are no exception. Below I attempt an objective analysis, and debunk claims that the decline in the official unemployment rate is not reflective of reality. However, I also show that any improvements should not necessarily be credited to the current administration. When reading, please remember that you can magnify a graph by clicking on it.

The most frequently-cited labor-market statistic is the unemployment rate, which is compiled by the Bureau of Labor Statistics (BLS). It is computed by dividing the number of unemployed workers with the sum of employed and unemployed workers. This sum is the labor force. To be considered unemployed, a jobless worker must have looked for employment four weeks prior to when they were surveyed. In May 2018, the U.S. unemployment rate, shown below on a monthly basis, stood at 3.8%, the lowest it has been since the late 1990s, and before that since the 1960s. Clearly, this is something to be happy about.

Some critics claim, however, that the low unemployment rate is an artifact. They argue that the drop is due to previously unemployed workers becoming discouraged and quitting their search. When this happens they stop being considered unemployed and are removed from the labor force. The critics argue that this explains the decline in the labor force participation rate, the percentage of the adult population that is either employed or looking for a job. The labor force participation rate on an annual basis is shown below, and we do observe a decline since 2000. Other critics argue that the official unemployment rate masks that some workers who are considered employed are working part-time because they are unable to find full-time employment. Finally, a third group suggests that while the unemployment rate may have gone down, the decline has not materialized in higher earnings for workers. Let's look at these claims in detail.

Is the decrease in the labor force participation rate due to unemployed workers becoming discouraged? The answer is no! It is mainly due to the fact that the baby-boomers continue to age and retire. When they retire, since they are neither employed nor looking for work, they are removed from the labor force. To show that this demographic trend is responsible for most of the decline in the labor force participation, I graph below the labor force participation rate of adults between the ages of 25 and 52. People in this age-range are typically not eligible for retirement. Hence, fluctuations in this rate are more likely to be driven by changes in employment opportunities than demographics. Here, too, we see a decline from 2000 until 2013. However, it is much smaller. More importantly, since 2014 the rate has been increasing, just as the unemployment rate has been falling. Therefore, it seems that the recent drop in the unemployment rate does reflect an improvement in labor market conditions, which has motivated prime-age workers to re-enter the labor force.

Alternatively, we can look at broader measures of unemployment, compiled by the BLS since 1994. These are shown below, on a monthly basis. The red line adds to the number of unemployed the number of discouraged workers. These are workers who had looked for employment in the 12-month period before they were surveyed but not in the past four weeks, and who gave up looking because they felt that their search was futile. Adding discouraged workers makes little difference, however, since it raises the joblessness rate by less than half of a percentage point relative to the official rate. The green line adds to the unemployed all marginally attached workers. These are workers who had looked for employment in the 12-month period before they were surveyed but stopped looking, regardless of the reason why they quit. Hence, this group includes discouraged workers but also workers who stopped looking because they decided to, say, go to college. Adding all marginally attached workers still makes little difference, as it raises the joblessness rate by less than one percentage point relative to the official rate. Finally, the purple line adds to the unemployed all marginally attached workers and those working part-time because they cannot find full time employment. Including underemployed workers to the pool does make a difference, since it raises the rate of joblessness by about four percentage points. Of course, one can argue that this rate overestimates unemployment, since those working part-time are at least partially employed. In any case, what matters for our purpose is that the four measures move together, so in terms of historical trends it does not matter much which one we look at. In fact, choosing the broadest measure suggests that the improvement in the labor market since 2010 has been greater, as this rate fell by more than nine percentage points while the official rate fell by about six. Visually, both the purple and the blue lines have been falling but the purple line has fallen faster, thus closing the gap with the blue line.

Finally, what about the claim that despite the drop in the unemployment rate there have been no gains in labor earnings? The graph below shows the real (adjusted for inflation) median weekly earnings of wage and salary workers. The median earnings is the earnings right in the middle between the lowest and the highest. We can see that contrary to that claim, median earnings have, in fact, been increasing since 2013, and are now higher than they were before the Great Recession. 

The conclusion that can be drawn once all these numbers are put together is that the current labor market is tight, and that during the past half decade there has been an improvement in labor market conditions both in terms of employment opportunities and in terms of earnings. At the same time,  the improvement began before the current administration took office. Hence, it is way too early to assess its footprint, as to do so we would need to examine whether the improvement accelerated or decelerated since President Trump took office, and we don't have enough data for that. Finally, while these numbers should make us happy, there are some signs of trouble. Specifically, there is some indication that since 2010 there exists a greater mismatch between the workers seeking employment and the jobs that are available to those workers. But more on that on the next post.

Varoufakis wrong and dangerous, this time about Italy

In an Op Ed published in The Guardian on May 28, 2018, Yanis Varoufakis claimed that:

While it is true that Italy is in serious need of reforms, those who blame the stagnation on domestic inefficiencies and corruption must explain why Italy grew so fast throughout the postwar period until it entered the eurozone. Was its government and polity more efficient and virtuous in the 1970s and 1980s? Hardly.

The singular reason for Italy’s woes is its membership of a terribly designed monetary union, the eurozone, in which the Italian economy cannot breathe and which consecutive German governments refuse to reform.

To understand why Varoufakis is wrong, let's look at exhibit #1 below. The blue line shows the annual percentage change in Italy's GDP per person, adjusted for inflation. The red line shows the same series for Germany after 1970, at which point the data become available. Keep in mind that Italy adopted the Euro on January 1, 1999.

The first observation one should make, for example by looking at the peaks of the blue line, is about what is missing. There is no sudden drop in Italy's economic growth after 1999 as would be the case if Varoufakis was correct. Instead, we see a gradual decline that has been taking place since the 1960s. And as an economist, Varoufakis should know that Italy's performance is consistent with what economic theory describes as "catch-up growth". Namely, countries whose incomes fall due to an unusual circumstance such as a war or natural disaster are expected to grow faster right after these circumstances are removed, and their growth to slow down as they close the gap with the rest of the world. This is exactly what we see in Italy, following the end of WWII.

The second observation is that Italy's growth rate traces Germany's pretty closely both before and after the adoption of the Euro, and until 2005. For the eight years from 2006 to 2014 Germany does outperform Italy but in 2015 the two growth rates converge again. Hence, there is little evidence to support the idea that Germany is, through the Euro, strangling Italy's performance, unless it is doing the same to itself. In fact, contrary to Varoufakis' claims, it does seem that Italy's lackluster performance relative to Germany's is of its own making. Specifically, the divergence between the two countries occurs in 2006, which was an election year followed by two political crises, in 2007 and 2008, and an unstable political environment since then.

Now that we have established that Varoufakis is wrong, let's look at why he is dangerous. Surprisingly, it is precisely because the Euro's architecture is faulty, as he claims. One of its biggest problems is that there is no institution that can act as a lender of last resort when a country-member faces a bank-run. No government in the world has the ability to pay off its debt within a few years. Typically, every government rolls its debt over by issuing new debt and using it to pay off the old. A phenomenon similar to a bank-run can occur if lenders become worried and refuse to lend to a government except perhaps at higher interest rates. If the government can issue its own currency, it could ameliorate the situation by using newly-issued currency to pay off the debt that is maturing. Doing so is no free lunch as it can lead to much higher inflation, but it is a solution. A country belonging to the Eurozone, however, does not have that option. It's only option is to borrow at the higher interest rates, which makes less likely that it will be able to service the interest on the new debt. As its debt becomes riskier, lenders may demand even higher interest rates, making the debt even riskier and so on. If this process spirals out of control, the country may go bankrupt as a result of a so-called self-fulfilling prophecy: a situation that happens simply because people expect it to happen. So yes, reforms are necessary, including the establishment of a mechanism that prevents such self-fulfilling prophesies, for example by offering a line of credit to countries that find themselves under pressure despite adhering to fiscal rules that ensure the sustainability of their debt in the long run. If that was what Varoufakis is making a fuss about, I would be happy to join him.

Unfortunately, through his narrative of us-southerners vs. them-Germans, Varoufakis seems to be advocating for higher levels of government spending even though government spending in the EU is significantly higher than in the US, and in Italy much higher than in Germany. Varoufakis seems to believe that economic prosperity is driven by the government. The problem is that Greece, among other countries, tried his remedy in the 1980s. To increase its spending, the Greek government ballooned its debt from about 25% of its GDP to about 100% of its GDP in less than a decade, and pursued a faster monetary expansion than the rest of Europe. And yet, Greece's economic growth during that time fell below that of the rest of Europe, reversing a pre-1980s trend. Greece's unemployment rate actually rose, and all the country had to show after the massive increase in spending was much higher inflation relative to the rest of Europe. By advocating for the tried and failed policies of the past, Varoufakis is reducing the chances of practical and possibly effective reform and, even worse, fueling populist movements that, if they have their way, may make today's woes seem like happy times.