Tuesday, July 26, 2016

The Misattribution Theory of 2016 Politics

Only half way through so far, the story of 2016 politics has already contained enough crazy plot twists to earn the "stranger than fiction" label. Last month, the UK shockingly voted to leave the EU, immediately sending its economy into a likely recession. This month, in the US, the Republican party nominated Donald Trump, a candidate who has advocated war crimes and suggested he might be open to leaving the World Trade Organization, abandoning NATO, and defaulting on the US debt, to name only a few of his radical ideas.

Instead of a page turner, the story of 2016 politics has turned out to be a horror story that has me afraid to turn the next page for fear of what might be next.

These unexpected events scream out for an explanation, and many have been suggested so far.

Rational Explanations

Those explanations I have seen have tried to describe the actions of voters in the UK and US as rational responses to current circumstances. They argue, for example, that voters have finally turned against the immigration and free trade policies that have left many in the working class behind.

Before considering whether immigration and free trade are the causes of these problems, let's first consider whether the claimed problems even exist at all.

Immigration


Immigration is deeply unpopular amongst US and UK conservatives who feel that the gains of immigrants are coming at their expense. While many careful studies have found that immigration raises the wages of native-born workers, let's ignore that question and instead ask "is immigration currently happening?"

One would naturally expect that a backlash against immigration in the current political cycle must be a response to an influx of new immigrants. In fact, over the past 7 years (going back to 2009) net migration from Mexico to the US has actually been negative. That is, more people have been moving from the US to Mexico than vice versa.

It's hard to see how the recent political backlash against immigration could be a rational response when immigration isn't even happening. At present, Donald Trump's wall would not keep Mexicans out but rather keep in those people trying to get into Mexico.

Crime


Trump has claimed that crime is out of control. In fact, despite recent tragedies making news headlines, crime rates are actually reaching historic lows.

Trump has also connected crime to immigration, infamously suggesting that Mexico is "sending us rapists". However, crime statistics show that immigrants are, if anything, less likely to commit violent crimes than native born citizens.


Wages


Talk of "stagnant wages" has been another mainstay of prior explanations of the current political climate. This explanation says that middle class workers are not seeing any of the gains that have accumulated to the rich in recent years. Hence, they are being left behind by a "rigged" system and are turning Donald Trump (and Bernie Sanders) for help.

Again, let's ignore the question of whether the system is rigged and instead focus on whether the described symptom, stagnant wages, really exists.

First, let me point out that the use of "wages" rather than "compensation" in these statistics is already suspicious. After all, we know that a large part of the compensation for many employees comes from employer-provided health insurance. We also know (1) that health insurance costs have continued to increase and (2) that more companies are providing health insurance due to the employer mandate in Obamacare. Hence, even if wages were flat, the above facts would strongly suggest that compensation has actually been increasing — it has simply been coming in the form of increased health insurance payments rather than increased wages.

Let's set that issue aside as well and instead ask whether the problem of stagnant wages actually exists. Here too, the answer appears to be no. At least for prime age (25-55) workers, wages have been steadily increasing throughout the recession. In the last year, wage increases have been on par with levels from before the Great Recession. When you consider the fact that interest rates are much lower at the moment — suggesting inflation is lower than before the Great Recession — these wage increases are even larger than they seem.

Again, it seems hard to argue that the current political climate is the result of stagnant wages, when wages are actually increasing at what was previously considered a solid rate.

Unemployment


Increasing wages are great for those with jobs, but perhaps we have unusually high unemployment? That does not appear to be the case. The current unemployment rate is below 5 percent, which is considered quite good.

Conservatives have instead pointed to the decline in labor force participation and suggested that the "real unemployment rate" is much higher. However, after accounting for an increase in retirees (Baby Boomers) and more people in school, that decline could mean, at the very most, 2.6 million more unemployed. Even that likely overestimate would only bring the unemployment rate up to 6.6 percent, which is comparable to the unemployment rate in the early 1990s and early 2000s. Those were certainly not roaring economies, but they didn't lead to political movements like we are seeing now.

Steel


Trump has complained about US steel companies being unable to compete with China due to (alleged) unfair trade practices. As this article explains, US steel companies are actually quite competitive. Over 70 percent of all steel used in the US is made in the US.


Manufacturing


Another common complaint is about the alleged decline in US manufacturing. However, US manufacturing output is actually near an all-time high. In fact, despite all the talk about China stealing manufacturing job from the US, right now, it's actually China losing manufacturing jobs to us!


Trade


Steel and manufacturing are two examples of industries in which many claim US companies have lost out due to free trade. As we just discussed, in neither case do the facts fit neatly with this story.

In general, economic theory tells us that free trade almost always benefits both parties. However, it also says that some (typically small) groups can be harmed by free trade in the short run. As a result, the US has government programs to redistribute some of the benefit of free trade to those who were harmed by it.

Perhaps those supporting Trump and his anti-trade platform are in the latter groups. The New York Times has done some investigating. In both cases they examined — where cities had fallen on hard times, blamed it on free trade, and are supporting Trump — they found that the declines were not due to trade at all. Businesses had moved to other parts of the US or simply lost out due to changing consumer preferences. In fact, in both cases, they found that those cities were actually being kept afloat partially by freed trade, and they would be in much worse shape without it.


Overall Economy


The evidence suggests that the average American is not oblivious to all of the improvements mentioned above. Consumer sentiment is high as are both the percentage of people expecting to be better off next year and also the percentage that say they are better off than last year. The improvement in the latter measure is as high as it has been since the late 1990s.


Everything is Amazing and Nobody is Happy


The economic expansion since the end of the Great Recession is already the fourth longest since 1850 and it shows no signs of stopping. Wages are increasing at a solid rate. Unemployment is somewhere between great and okay depending on how you measure it. Crime is down. Even race relations have been improving.

Despite all of this, there remains deep unhappiness and anxiety. The same article linked above points out that, while consumer sentiment is high, there continues to be low satisfaction with "where the country is going". The latter trend matches the spirit behind Brexit and Trump, but it doesn't seem to be rationally explainable by most commonly cited causes discussed above. We will have to look elsewhere.

Global Trends

One common thread between Brexit and Trump is the rise of nationalism. Support for both is based on dislike of immigration and includes calls to put the UK/US first.

The US and UK economies have gone through similar journeys in recent years, which would seem to support economic explanations for these dramatic political events. However, any explanation for these trends should also need to explain similar events happening around the globe in countries that do not match this pattern.

For example, as Scott Sumner has asked, "Why is nationalism on the rise in China, where real wages... have been soaring at double-digit rates in recent decades?" And "why is right wing populism on the rise in Austrlia, despite the economy performing brilliantly over the past quarter century?"

We will have to look elsewhere for an explanation that fully fits these facts.

The Misattribution Theory

One common problem for all of the claimed causes listed above is that they seek to find explanations that are rational, that describe these political events as a reasonable response to the situation voters find themselves in. However, we know that all humans are irrational (at least partly).

Indeed, as Bryan Caplan has pointed out, it is generally accepted that participants in the stock market, where billions of dollars are on the line, often act in a clearly irrational manner, so why would we expect voters to behave in a way that is fully rational?

Instead, we will look for an explanation amongst the long list of experimentally demonstrated cognitive biases. For me, the most compelling explanation of recent political events comes from the psychological concept called "misattribution".

Misattribution


In 1974, Dutton and Arons performed the following experiment. A female interviewer would ask questions of men who had just walked across a long suspensions bridge. Afterward, she would give them her phone number in case they had further questions. The experimenters anticipated that those men who found the interviewer physically attractive would be more likely to call back.

The experiment considered two different groups of men. For those in the first group, the interview took place immediately after they crossed the bridge, while their hearts were still beating rapidly from the excitement / fear of being a long way above the ground and their breating was still rapid from the exercise. For those in the second group, the interview took place after the men had a chance to rest.

The experimenters hypothesized that, since rapid heart rate and breathing are signs of physical arousal, the men who were interviewed immediately might confuse those bodily signals for evidence that they were aroused by the female interviewer. As predicted, more of the men interviewed immediately choose to call the interviewer afterward.

In short, the idea of misattribution is that our feelings can influence us in ways of which we are unaware.

A Theory of 2016 Politics


My hypothesis is that recent political trends are the result of misattribution of the feelings of anxiety caused by rapid change in the world in which we live:

  • Economic Change: The rise of automation is making many jobs obsolete. In fact, for both of the examples cited above, steel and manufacturing, the higher-than-ever output of these industries achieved with lower-than-ever employment is the result of automation. China has not been stealing manufacturing jobs (at least not more than temporarily), rather these jobs are being automated away.
  • Technological Change: The aforementioned automation is technological change, but it is far from the only example. The rise of AI is likely to affect the jobs of those in industries beyond manufacturing and materials. Self-driving cars and trucks, while saving many lives, also raise the specter of job losses in transportation.
  • Social Change: While gay marriage has been a political issue for over a decade, the move toward acceptance of transgender citizens is a recent change. Likewise, the recent videos of police shootings of African-americans are causing acceptance among many of the long claimed bias of police against them.
Change is not the only cause of anxiety, however. The threat of terrorism is also a cause. Indeed, the very goal of terrorism is to cause such feelings. Despite the fact that the average person is still more likely to accidentally drown in a pool than be killed in a terrorist attack, the threat of such attacks is ever present in many of our minds.

All of these causes lead to deep feelings of anxiety about our jobs, our society, and our very lives. This response is entirely rational.

I suspect, however, that many are feeling this anxiety and, like the men being interviewed after walking across the bridge, misattributing its cause to explanations in front of them. If we believe (or are told) that China is stealing our manufacturing jobs, these anxious feelings are bodily signals confirming that we are right to be concerned about China. If we believe (or are told) that immigrants are coming to America and causing crime, the anxious feelings from our bodies tell our minds that those fears are real and reasonable.

The evidence above demonstrated, however, that these fears are not reasonable. Wages are not stagnant. Unemployment is not high. Crime is not increasing. Hence, if Americans have a bad feeling about "where the country is heading", then we have to question the true source of that feeling.

Unlike the economic explanations, the misattribution theory fits with examples beyond the US and UK. The types of change described above are not occuring only in the US and UK but all across the world. For example, while the economies in China and Australia have performed very well in recent decades, they are experiencing the effects of automation and changing technology. Thus, the misattribution theory offers an explanation for the rise of nationalism in places around the globe.

Conclusion

While most analyses of the visible trends in 2016 politics take voters stated beliefs at face value, it is reasonable to be suspicious when voters blame their problems on their usual bogeymen. On the right, they blame foreigners, whether residing in the US (immigrants) or in their native countries, as well as those of different religious and ethnic backgrounds (e.g., Muslims). On the left, they blame corporations, so much so that a candidate simply willing to accept a no-strings-attached donation to their campaign is then viewed as untrustworthy.

While others may claim that these are recent trends, in response to the current social and economic conditions, that is simply not the case. Survey data shows that the median voter has, for a long time, been both a nationalist (in line with fears on the right) and a socialist (in line with fears on the left).

Distrust of foreigners and/or corporations has been a consistent tradition in the US (and I suspect in the UK as well), making it fruitless to try to view them as responses to recent changes in economic and social conditions.

It is certainly possible that recent changes have caused an increase in enthusiasm for these views, but as I said above, we should be naturally suspicious when voters respond to changes by blaming the usual bogeymen, even if they do so with increased enthusiasm. The median voter has demonstrated a consistent tendency to blame problems on these entities, making their claims that "this time it's really true" hard to accept.

This is not to say that we should be skeptical of the median voter's testimony all together. Each voter is, of course, the primary expert on their own condition, so we should take seriously their claims about the state of their own finances, for example. However, while there are certainly parts of the country that remain in bad economic and social condition, we saw above that the average voter admits that their own economic condition has improved more in the last year than at any time since the 1990s.

It seems to me that misattribution is a much better fit to these data. The median voter is anxious about a great many changes taking place in society today. I feel these same anxieties myself. But the tendency to explain these anxieties as a rational response to worsening conditions is at odds with the facts, and the claims that these (nonexistent) worsening conditions are the fault of the usual bogeymen should be viewed with deep suspicion.

The world is in fact getting better in ways that few would have imagined only decades ago. The world is now safer, richer, and healthier than ever before in human history. I agree with Mark Perry that the drop in world poverty to below 5% worldwide represents the greatest human achievement in our history. Free trade and globalization have lifted over 1 billion people out of poverty in the last few decades — an achievement that cannot be understated.

While the rapid economic, technological, and social change is a cause of anxiety for many, we must fight against our natural inclination to turn anxiety into fear and anger toward those we distrust. As the facts above demonstrate, the vast majority of all the change we have seen has been for the better. The future ahead is safer still, with self-driving cars; healthier still, with continued progress in the fight against cancer and heart disease; and richer still, with rise of AI making legitimate the prospect that many will have no need to work at all. The only thing that can derail that progress is giving in to our fears and turning on each other.

Saturday, June 11, 2016

The Anti-Establishment Cycle

One common theme between the surprising successes of both Donald Trump and Bernie Sanders in this political campaign is an anti-establishment, anti-Washington sentiment. Far from being a new idea, it seems that we see a similar anti-establishment rise every few elections. While there are likely multiple causes of anti-establishment sentiment, I have one theory for why it never seems to go away, even when the old politicians are all thrown out and replaced with fresh new faces.

Let me explain this with an example: the plan to build a wall along the Mexican border. This has a been a key plank of Trump's campaign; however, it is certainly not an idea that he invented. The US government started building a security fence along the southern border during George W. Bush's presidency and continued it under Barrack Obama, yet this continues to be a key political issue in the current campaign because many voters perceive illegal immigration to still be a problem (whether they are correct about that is another matter).

From the voter's perspective, George W. Bush promised to stop illegal immigration and didn't deliver. The Republican's in control of Congress during Obama's presidency promised the same and also didn't deliver. What can explain this? The most obvious explanation is that these politicians aren't up to the job or, even worse, they don't actually want to solve the problem. Perhaps they are being paid off by special interests that want illegal immigration to continue!

In the case of the wall, none of these explanations is a good fit. As John Oliver explained, the real problems are that (1) building a wall turns out to be a lot harder than anyone thought and (2) the wall doesn't actually fix most of the problem. (Most illegal immigrants enter the country legally and then over-stay their visa.)

Let me abstract from this example to a more general description of what I think happens in many cases. First, the voters have a problem they want solved. Politicians come along to say that the problem has an easy answer but opponent's don't want to solve the problem because they're corrupt (or under the thumb of special interests), so the voters should give them power instead. Once elected, the politicians discover that the problem is a lot harder than they thought, but they don't want to admit that (because it would show their ineptitude), so they continue to blame their opponents. After a while, voters reach for the simplest explanation: these politicians must be corrupt too. So they throw them out and elect new politicians with a new easy answer.

In our example, the problem is illegal immigration and the easy answer is building a security fence. That didn't solve illegal immigration, so voters start to suspect that Republican's in Congress are incompetent or worse. Trump comes along with a new easy answer (replace the fence with a huge wall!) and explains that those who say it won't work actually want illegal immigration to continue.

Here's another example: on the Democratic side, Bernie Sanders says he can solve "too big too fail" by breaking up the big banks (the easy answer). The only reason that others won't implement his easy answer is that they're corrupt: they're being paid off by the big banks. In reality, the problem is far more complicated than this. For example, one of the reasons that we have a few big banks is that banks are so highly regulated. (Read the annual reports of M&T Bank to see how a smaller regional bank copes with regulations like Annunzio-Wylie, Sarbanes-Oxley, Dodd Frank, etc. and why it is more expensive for them than the larger banks.) More regulation of the banks is likely to make the big banks even bigger.

These are far from the only examples.

Why won't politicians raise the minimum wage in order to help poorer workers? It must be because they don't care about them. Again, the issue turns out to be more complicated. Until the last two decades, pretty much all economists believed that raising the minimum wage would help some workers but hurt others by taking their jobs away. Some more recent studies (e.g., Card and Krueger) found that this didn't happen in examples that they looked at, but they instead found businesses paid the higher wage by increasing prices. And since businesses that pay the minimum wage more often have poorer customers, you have the same problem of helping some poorer people and hurting others. (In this case, I have to say that I think there is an easy answer: the government should just give poor people money. However, the point remains about the minimum wage as the easy answer.)

Why won't politicians raise taxes to pay for more services? It must be because rich people are paying them off. In this case, the problem is that the revenue maximizing tax rate is typically much less than 100%. There have been some empirical studies on this (none of them great), but they suggest that the maximizing rate is around where tax rates are now. And in fact, despite having had a highest tax rate as high as 90% at some points in recent history, tax receipts as a percentage of GDP have only been higher than they are now at two points in history (2000 and 1945).

Why won't politicians cut taxes? It must be because they hate rich people. Nearly all of the Republican candidates in this campaign want to drop tax rates to around 20% and they claim that this will actually increase tax revenue either by increasing growth or because the current rates are above the revenue maximizing rate. As I noted above, the second claim is very likely false. As for the former, it too rests on shaky ground.

I could go on. Indeed, I actually think that the vast majority of all political problems are much more complicated than politicians want to admit. At the end of the day, "it's complicated" doesn't win many votes. Voters would rather believe that there are easy solutions,whose implementation is only being prevented by corruption. But when those easy solutions don't pan out, we can expect them to reach the only reasonable conclusion, that corruption has spread, and start the next wave of anti-establishment sentiment.

Friday, June 10, 2016

The One Claim From Which Trump Cannot Back Away

To say that this year's race for the Republican presidential nomination has been strange is of course an enormous understatement. The presumptive nominee, Donald Trump, as others have noted, does not even hold conservative views on most of the issues that Republican's have typically cared about. On those issues which Trump cares about, he holds views well outside of the norm, endorsing war crimes, religious persecution, mass deportations, and torture.

As a political moderate, I certainly do not agree with any of those positions (indeed, I find them morally repugnant). However, the purpose of this post is not to criticize any of the listed views of Donald Trump. Instead, I want to discuss one of Trump's outlandish positions that is unlike the others.

What distinguishes the position I have in mind is that it is one from which Trump cannot ever back away effectively. That matters because mainstream conservatives appear to believe that Trump will soon moderate his stances on these issues in order to make himself more electable to moderates (like myself). Whether he will do so remains to be seen, but the Republican establishment appears hopeful that Trump will repudiate his endorsements of those positions that are well outside the norms of American political beliefs, including all of those listed above (war crimes, religious persecution, torture, etc.).

What distinguishes the position I have in mind is that it is not one where Trump can undo the damage simply by backing away. For issues like war crimes or torture, if Trump changes position, becomes president, and never carries out his earlier proposals, you can argue that no real harm was done. No crimes were committed. No one was tortured. While his statements were disturbing, the damage pales in comparison to that of actually carrying out these ideas. So if he were to back away and never follow through, one could argue that there was little harm done.

For one of his proposals, however, I will argue that the damage is already done merely by it's suggestion. That proposal is the idea of defaulting on the US debt.

Before I begin in earnest, let me note that Trump was careful to avoid the word "default". That word has a negative connotations. Instead, he suggested that we "re-negotiate" our debt. However, the latter is something that occurs in default, so he is simply skipping mention of the step that sounds bad, despite the fact that it occurs before the part he did mention.

At present, US debt has the highest possible credit rating. It is referred to as "risk-free" because, while not literally risk free, it is believed to have less risk than any other financial asset. One key reason for that belief is that US government, regardless of which party is in control, has continued to publicly state that it would not, under any circumstances, default on that debt. It is generally believed that, in the worst case, rather than default, the US would increase taxes on its citizens in order to make good on the payments it has promised to debt holders.

The benefit of the belief that US debt is "risk-free" is that there is a large market for that debt. A great many individuals, corporations, and governments want to purchase our debt because they wish to avoid risk. As a result, the US is able to issue debt (i.e., sell bonds) at very low prices, meaning that the interest payments for the US are smaller than those of countries whose debt is seen as riskier.

The effects of low interest rates are probably familiar to anyone who owns a home (or who finances the purchase of a car): the lower the interest rate, the lower the monthly payment. Hence, the US has lower monthly payments per dollar of debt than most other countries in the world, which means we get to keep more of our money to spend on services for US citizens.

The key part of this for our discussion is that the interest rate on US debt is driven by perceived risk. The US pays a low rate because investors believe that debt is risk-free. If they start to worry that there is risk of default (even a little), then rates start to increase.

That may not sound so bad, but worries about default often become a self-fulfilling prophecy. When fewer investors want to buy, interest rates go up, which means the US has to pay more, which increases the risk of default.

Cosmo: ... Everything in this world, including money, operates not on reality... 
Bishop: but on the perception of reality.
Cosmo: Posit: people think a bank might be financially shaky.
Bishop: Consequence: people start to withdraw their money.
Cosmo: Result: pretty soon it is financially shaky.
-- Sneakers (1992) 
If you want a real world example of this, see the crisis of confidence in Greek debt that led to its default in 2012.

While Trump may furiously back-pedal on his suggestion of defaulting on US debt, as Cosmo noted in Sneakers, financial markets operate not on reality but on the perception of reality. If worldwide investors believe that there is even a small increase in real risk of default due to Trump becoming president, then interest rates will start to rise if he is elected. (Indeed, even Trump's nomination may cause an increase in perceived risk of default.)

That perception has real world consequences. In the short run, our debt will be harder to pay off, which means there will be less money remaining for services benefiting US citizens. In the long run, the risk of actual default will undoubtedly increase as well.

As hard as Trump may try to disavow this proposal, his one-time suggestion of default is a permanent stain on his candidacy that I do not believe can ever be fully erased.