This post is part of an occasional ongoing series about the forces driving political partisanship in the United States, based on original research conducted by ClearerThinking.
The state and direction of the economy plays a major role in any American presidential campaign. Typically, the debate revolves around how to maintain momentum in flush times, or how to kickstart the economy if it's moving sluggishly or in recession. The 2016 campaign was an aberration in that sense — the primary economic question was not "What should we do regarding the economy?" Rather, it was "How is the economy even doing right now?"
So how is the economy doing right now? Our research into the attitudes of Clinton and Trump supporters at the end of the 2016 electoral cycle found substantial disagreement on this question. When surveyed, 28% of Trump voters agreed with the notion that "America is currently performing well economically," while 58% of Clinton voters agreed with the same proposition. By contrast, 61% of Trump voters disagreed, compared to just 24% of Clinton voters. (In the graphic below, red bars represent Trump voters, and blue bars represent Clinton voters; the numbers on the X-axis indicate the intensity of respondents' feelings on the subject.)
It's tempting to ascribe these disparate attitudes to social media-abetted political echo chamber effects or to the proliferation of highly partisan news outlets, and those factors almost certainly do play a role in this phenomenon. But there's another, more basic element at work here that's just as important to understand: the U.S. economy is a highly complex system whose overall health is hard to evaluate, and the metrics used to measure it often tell conflicting stories.
To illustrate this point, let's quickly look at commonly-used economic indicators and see what they say about the American economy's performance over the past decade or so. Perhaps the most famous and frequently covered of these indicators is the U.S. stock market, whose movements generally reflect the behavior of the entire economy, at least up to a point. By this metric, the U.S. economy has done quite well for itself since hitting its post-financial crisis low point in 2009. The graph below charts the value of the S&P 500 — an American market index that closely mirrors the economy as a whole — and shows a consistent upward trend since the 2009 crash.
But the movements of stock markets don't fully capture the health of the economy. Another significant economic indicator is the unemployment rate. This indicator is of special interest to politicians, as job anxiety is a strong driver of voter behavior. Here, the picture looks more ambiguous, but still tracks closely with the positive trend found in the stock markets since the financial crisis. The U3 unemployment rate — which is the official rate used by the government — fell from just shy 10% to its current position around 5% over the course of the Obama administration. By contrast, the U6 unemployment rate — which uses a broader definition of unemployment than the U3 rate, including discouraged and underemployed workers — peaked at close to 17% in 2010, but has fallen to just below 10% since. (The U6 unemployment rate is arguably a clearer measure of economic anxiety due to its broader definition, but each has its proponents.)
These two indicators have moved in conjunction with the trends reflected by the S&P 500 during the same period, but they tell different stories about how much unemployment Americans are facing and about the economy's real overall health.
Judging by these two popular indicators, the American economy is doing great. Stock prices are soaring to unprecedented heights, and while the job market looks different depending on which measure you use, it's very clearly been moving in the right direction for years.
But that's not the whole story. While stocks have risen and unemployment has eased, the value of actually having a job — specifically, "real" (i.e. inflation-adjusted) wages — has not meaningfully ticked upwards. In fact, the average American hourly wage has been stagnant for decades, rising from $19.18 to $20.67 in inflation-adjusted terms between 1964 and 2014. (It's worth noting that this figure includes only cash payment and doesn't account for health insurance or other benefits.):
The lack of movement in this metric is especially significant, because the prices for essential goods and services have not remained constant over the same period.
To get a sense of why this fact is so important, consider the behavior of student loan debt and healthcare costs in recent years. As you can see from the graph below, the average annual healthcare expense per capita rose from $3,932.45 to $7,282.32 between 2000 and 2012 – an 85% increase. You would expect this kind of cost increase to accompany a massive increase in the quality or value of the care provided, but it seems extremely unlikely that 2012's healthcare really outperformed 2000's by 85% for most Americans. And remember that wages were stagnant during that period — the average American therefore devoted a substantially larger portion of her total earnings to healthcare costs in 2012 than she did in 2000. That's not good news.
The student loan debt picture is similar, but perhaps even more alarming. In the decade between 2006 and 2016, the total national student loan burden rose from $481 billion to $1.35 trillion – a 181% increase! It's hard to explain this increase in terms of population growth, inflation, or even educational quality. (In fact, it's not obvious that education quality has gone up at all over this period.) And again, wage growth was flat during this period; higher education simply became much harder for average Americans to pay for during this period, which has troubling implications for social mobility and long-term economic health.
You can see a similar pattern in K-12 academic spending in this graph, though it took a much longer time (some 35 years) for this sector to see the kind of percentage cost increases that student loan debt saw in a decade:
This phenomenon — wherein essential services become rapidly more expensive without substantial quality improvements, even as wages remain flat — is known among certain economists as "cost disease," and it's apparent in other areas of public life as well, such as infrastructure costs and housing expenses. While the stock market and unemployment rates are important indicators that have painted a relatively rosy picture of the American economy, the combination of stagnant wages and cost disease have placed substantial new pressures on American consumers. For a lengthy and thorough look at the question of cost disease, we recommend checking out this excellent post on Slate Star Codex.
This complexity is important to keep in mind when considering the divergent impressions of the economy held by each side of the American political spectrum. These impressions vary not only because of ideological siloing and partisanship, but also because the data itself is deeply conflicted — the considerations we've discussed here are really just the tip of the iceberg when it comes to understanding the current state of the economy. To make matters even more challenging, people tend to base their impressions of the national economy on the local economy wherever they live. Economic conditions vary wildly from place to place — and since political preferences are strongly correlated with region, it's easy to see how Trump voters and Clinton voters might end up with incompatible impressions of America's economic health, even though neither is really wrong.