Listen to this article
Give us your feedback Thank you for your feedback.
What do you think?
At least two big questions about tax are convulsing Washington DC this week. One is what type of tax reform bill might eventually squeak its way through Congress. The other is whether this bill will offer sweeteners to maintain voter support for President Donald Trump.
But as the pundits keep crunching the numbers — and furiously disagreeing about the economics behind the putative bill — there is a third question that also needs to be asked: will the future economic impact of any tax reform really influence how voters feel?
In some senses, this might seem a peculiar question to ask. After all, economists and political strategists have tended to assume in recent years that people vote to a large extent with their wallets. Or, as advisers to Bill Clinton, then a presidential candidate, observed in 1992, politics is about “the economy, stupid”.
But this might now be wrong. Economists from Chicago’s Booth School of Business and Princeton University have recently completed the first in-depth study of economic sentiment, political affiliation and household spending in the Trump era. And this has a startling message that matters for the current tax debate: namely that in our post-truth Trumpian world, politics seems to be driving economic sentiment, as much as the other way round.
Or, as the paper (by Amir Sufi, Atif Mian and Nasim Khoshkhou) says: “Partisan bias is exerting a stronger influence on economic expectations over time . . .”
To reach this conclusion, the authors crunched long-run extensive data sets from the University of Michigan consumer survey and Gallup opinion polls, broken down by postcodes. This created a detailed picture of monthly changes in household sentiment and political views. They then compared this to detailed state- and country-level data on tangible economic trends.
This shows that ever since the data started to be collected (five decades ago), political affiliation has always exerted some influence on sentiment. That is unsurprising: if the Democrats have just won an election, say, it might seem natural for Democrats to feel more cheerful.
But what is startling is that this pattern has recently intensified sharply. Most notably, the data show that this year the link between party sentiment and economic optimism (or pessimism) has risen four fold, compared to the era of George W Bush. And if you look only at the optimistic Trump voters, you can see that the rise in optimism was “three to four times larger than that of Barack Obama’s supporters after the 2008 and 2012 elections” and six times larger than the swing seen when Mr Bush was elected in 2000, the authors say.
Moreover, this pattern is not just related to abstract national economic trends. Republicans and Democrats have starkly different ideas about what is happening to their finances and local economies — even in the same postcode.
What does this mean? One implication is that it shows the folly of assuming that voters are as rational as economists like to think. Another implication is that “soft” economic surveys can be unreliable guides to what is really happening in the economy. In fact this study suggests that there is relatively little evidence that surging economic confidence really leads to higher household spending. That reflects a bigger macro trend: although US consumer confidence has leapt to a 17-year high in recent months, consumer spending is only rising at an annualised rate of 2.4 per cent, slower than might have been expected on historical trends.
But there is a third implication, too: economic surveys might offer a better guide to how voters will vote than numbers. If so, that has implications for how pundits interpret the current Washington dramas. Democrats hope they will see a surge of support if the tax reforms end up hurting middle-class voters.
But if Mr Trump manages to persuade voters that their team is “winning” in a cultural sense — say, by stoking the culture wars via his Twitter account — then that may drive economic perceptions. And from what I have seen in my own conversations with voters, this is precisely what is going on. The genius of Mr Trump is that he has recognised that tribal sentiment matters as much as statistics. This is uncharted territory for economists.
Of course, this may just be down to a time-lag effect. It may also be merely a function of an economic recovery: if the economy crumbles, the pattern could shift. But until then, that old Clinton campaign adage seems only half-right. Democrats be warned.