The Economy of 2016 Weighs on Elections in 2018 and 2020
Voting patterns followed the contours of the lumpy and uneven recovery from the recession.
Bloomberg, June 15, 2017
One of the conundrums of the recovery from the financial crisis is how varied it is and the diverse perceptions of it by different people in different parts of the U.S.
All of this reminds me of John Godfrey Saxe’s wonderful poem, “The Blind Men and the Elephant.” In it, six blind men each touch and describe a different part of the animal. Their incomplete perspective makes it impossible to grasp the whole.
The elephant analogy applies to a U.S. economy, where regional components are very different from, and often bear little resemblance to, one another. Attempts to draw broad conclusions from a single locale or region will inevitably lead one astray.
Regular readers will recall that I have looked at these various differences before, usually in my effort to describe why for so many Americans it feels like the recession never ended. Much of the commentary has been about politics driving sentiment about the economy.
But I’m curious about the opposite causal relationship: How does your local economy affect national politics and state-level voting? It is through this lens that I am trying to understand the red-blue divide that is becoming so stark in America, turning the country into rival camps that neither recognize nor understand one another.
Before we consider that, let’s review some prior discussions: We have noted how strikingly uneven the recovery has been across several different facets of the economy. Housing, for example, has recovered strongly in some areas while it remains in the doldrums in others. Regional differences are stark, with rural areas that once depended on manufacturing in decline while those urban areas with information-services industries are thriving. The same is true by economic strata, based on whether you are in the top 1 percent or the top 0.1 percent. The top percentiles have garnered an inordinate share of all income gains since the recession ended in 2009.
No doubt, the slow recovery from the credit crisis deserves some blame. When we look at the numbers since 2009, we see a steady but gradual improvement. Unemployment, which peaked at 10 percent, has been more than halved, and now is well below 5 percent. In the eight years before the November election, some 12 million new jobs were added. Employers are now complaining of a shortage of workers in some industries. Job openings, especially for skilled workers, can go unfilled for long periods of time. Wages, which stagnated or fell in real terms for half a decade, only now are rising faster than inflation.
But it isn’t all rosy, as several important economic data points reveal: Gross domestic product growth is barely above 2 percent, hardly enough to generate sustained and broad optimism among consumers. Most forms of credit remain tight, especially for mortgages. Imagine the frustration of a potential homebuyer who can’t take advantage of the lowest rates in modern history. Although overall inflation is modest, important items such as health care, housing and education have skyrocketed. And while job growth has been steady, many of those are in low-paying service industries. The gig economy, touted for its growth potential, leaves lots of people underemployed. This comes on top of three decades of little-changed inflation-adjusted income for most households.
How happy you are is likely dependent upon how some of the factors affect you.
I was discussing this question with statistician Salil Mehta, who teaches at the Columbia School of Professional Studies’ Applied Analytics program, and was the director of analytics in the federal government’s $700 billion bank bailout plan, the Troubled Asset Relief Program.
Mehta looked at the voting patterns across the country, to try to find data that explained the surprising election outcome. It may have been a shocker to many people, but not to Mehta1 — he noted that many pollsters were underestimating Donald Trump’s chances of victory, based on some of the data below.
The economic data points with potential electoral consequences fall into a few broad types. The first are those red states that voted for Trump where growth and GDP were below the national rate. Mehta explains: “Republican states had a collective GDP per capita (chained to 2009 dollars) of $45,000, which is fine relative to some other nations; but for Democratic states it was nearly 25 percent higher at $57,000.” Those doing better economically tended to vote status quo/incumbent, while those doing worse tended to vote for the change/challenger candidate. Along the same lines, Mehta noted that “Republican states saw only a 2.7 percentage point improvement in unemployment, since the last election, while Democrat states saw a 3.6 percentage point improvement.”
Where things get even more interesting is Mehta’s analysis of what he called the forgotten states. These states have certain things in common: 1) below-average GDP per capita; 2) below-average unemployment improvement; and 3) they previously voted for Obama. There are six states that meet these qualifications — Florida, Michigan, Wisconsin, Ohio, Pennsylvania and Iowa. All went for Trump last year.
Many people are still coming to grips with the election of 2016. But, in retrospect, the signs were there for all to see. I bring this up because it’s something for incumbents and challengers to keep in mind when looking ahead to 2018 and 2020. The U.S. is not one economy, but many, and like politics, it’s intensely local. How the different regions, industries and demographic group are faring will play critical roles in future electoral outcomes.
1. Mehta is the sort of numbers geek that lives for data analysis and has little use for partisan politics.