Archive for the Category Cognitive illusions


This post is not about Scott Alexander and Trump

Scott Alexander has a new post out about Trump, and it’s everything this blog is not. (I.e. it’s intelligent and non-hysterical)

A commenter on here the other day quoted an Atlantic article complaining that “The press takes [Trump] literally, but not seriously; his supporters take him seriously, but not literally”. Well, count me in that second group. I don’t think he’s literal. I think when he talks about building a wall and keeping out Muslims, he’s metaphorically saying “I’m going to fight for you, the real Americans”. When he talks about tariffs and trade deals, he’s metaphorically saying “I’m going to fight for you, the real Americans”. Fine. But neither of those two things are a plan. The problem with getting every American a job isn’t that nobody has been fighting for them, the problem with getting every American a job is that getting 100% employment in a modern economy is a really hard problem.

Donald Trump not only has no solution to that problem, he doesn’t even understand the question. He lives in a world where there is no such thing as intelligence, only loyalty. If we haven’t solved all of our problems yet, it’s because the Department of Problem-Solving was insufficiently loyal, and didn’t try hard enough. His only promise is to fill that department with loyal people who really want the problem solved.

I’ve never been fully comfortable with the Left because I feel like they often make the same error – the only reason there’s still poverty is because the corporate-run government is full of traitors who refuse to make the completely great, no-downsides policy of raising the minimum wage. One of the right’s great redeeming feature has been an awareness of these kinds of tradeoffs. But this election, it’s Hillary who sounds restrained and realistic, and Trump who wants the moon on a silver platter (“It will be the best moon you’ve ever seen. And the silver platter is going to be yuuuuuge!”)

Read the whole thing.

But this post is not about Scott Alexander and Trump; it’s about cognitive illusions involving numbers.  This appeared in the same post:

If Trump fails, then the situation is – much the same, really, but conservatives can at least get started right now picking up the pieces instead of having to wait four years. There’s a fundamental problem, which is that about 30% of the US population is religious poor southern whites who are generally not very educated, mostly not involved in US intellectual life, but form the biggest and most solid voting bloc in the country. If you try to form two parties with 50% of the vote each, then whichever party gets the religious poor southern whites is going to be dominated by them and end up vulnerable to populism. Since the religious poor southern whites are conservative, that’s always going to be the conservative party’s cross to bear and conservatism is always going to be less intellectual than liberalism in this country. I don’t know how to solve this. But there have been previous incarnations of American conservatism that have been better at dealing with the problem than this one, and maybe if Trumpism gets decisively defeated it will encourage people to work on the problem.

This isn’t accurate.  I don’t know the correct figure, but the following explains how my brain works.  Without looking up any of the numbers, here’s what I’d guess:

1.  About 14% of Americans are poor.

2.  About 7% of Americans are poor whites.  The rest are poor blacks, Latinos, Asians and Native Americans.

3.  About 3% (at most) of Americans are poor white southerners.  The rest of the poor whites live in the East, Midwest, or West.

4.  About 2% of Americans are religious poor southern whites.

There’s a big difference between 2% and 30%, and this affects Scott’s argument. There’s a tendency (which I sometimes fall into) of assuming that the most distinctive characteristic of a candidate’s supporters is also the majority of the candidate’s supporters.  The whites of West Virginia form a more distinctive part of Trump’s base than the whites of affluent suburbs in Southern California.  But Trump will win far more votes in suburban Southern California. He’ll get more votes from college grads than high school dropouts, even while being the first GOP candidate in ages to (narrowly) lose the college vote. I have commenters who are extremely intelligent, and plan to vote for Trump.

A conservative might argue that the 2% of religious poor southern white voters who are mindlessly supporting the right is offset by an equal or greater number of poor black and Hispanic voters who mindlessly support the left.  Those two groups don’t decide elections.  If Trump wins, it will be because millions of highly educated professionals also voted for him.  Let’s not blame “stupid” poor people.

PS.  Even if you define “poor” more generously, assuming that Scott meant to also include lower middle class workers, you still don’t get past 10% of the electorate, probably not even 7%.

PPS.  Keep in mind that “the South” includes millions of affluent whites in the vast Texas triangle of metro areas, and Florida (which has more people than New York State), and the triangle region of North Carolina, and the Virginia suburbs of DC, and the affluent Atlanta suburbs and lots of other areas like that. I’d guess there are more “atypical southern areas” than typical southern areas.

PPPS.  Tom sent me a great youtube of Sam Harris discussing Trump’s mind. Trump’s suggestion that we should have taken Iraq’s oil has been weirdly overlooked. If it’s not the most disgusting thing said by a presidential candidate since the Civil War, it’s surely in the top 5.



The asset markets keep me from going insane

Over at Econlog I have a new post pointing out that negative IOR has had an unquestionably positive impact on asset prices, and yet much of the business press claims exactly the opposite.  This confused thinking makes it more likely that central banks will adopt bad policies in the future.

A similar problem occurred in late 2007 and early 2008, when the media adopted a Keynesian approach to monetary analysis, instead of a monetarist approach.

From August 2007 to May 2008, the Fed repeatedly cut interest rates, from 5.25% to 2.0%. The media treated this as an expansionary monetary policy, even though it was clearly exactly the opposite.  The monetary base did not change, and falling interest rates are actually contractionary when the money supply is stable.  Indeed it’s a miracle the economy didn’t do even worse.  NGDP growth slowed sharply, and I surprised there wasn’t an outright decline.

Here’s the monetarist approach:

NGDP = MB*(Base velocity), where V is positively related to nominal interest rates.

Thus if you cut interest rates without increasing the money supply, then V falls and policy becomes more contractionary. It’s monetary economics 101, but almost everyone seems to have forgotten this simple point.  Market’s responded favorably to larger than expected rate cuts, because they implied a bigger than expected boost to the monetary base, on that day.  But over time the base did not increase at all; the rate cuts were merely enough to keep it from falling.  So do more!!

Because the media wrongly thought money was getting easier, they became (wrongly) pessimistic about the efficacy of monetary policy, which led to President Bush’s failed fiscal stimulus of May 2008.  There is no economic model where Bush’s policy would be effective.  Interest rates were above zero, so monetary offset was fully applicable.  The Fed responded by putting rate cuts on hold, which drove the economy right off the cliff after June 2008.  By September the tight money caused Lehman to fail, as its balance sheet was highly leveraged, and exposed to asset price declines triggered by falling NGDP expectations.  Yet even Ben Bernanke inexplicably endorsed Bush’s tax rebates, even though there is no logical reason for him to have done so.  If more stimulus was needed in May, then the stance of monetary policy should have been more expansionary.  So do more!!

Bush’s policy was a lump sum tax rebate, which doesn’t even work on the supply-side.  And unlike more government spending, there isn’t even a New Keynesian argument that fiscal stimulus might boost aggregate supply by making people work harder.  It was really dumb policy, there’s nothing more to say.

So because the media and many economists wrongly though money was loose, we ended up with really bad macro policy. The recent backing away from additional negative IOR is more of the same.  Just as markets responded to unexpected rate cuts in late 2007 as if they were highly expansionary, markets responded to negative IOR in Europe and Japan as if it is expansionary.  But over longer periods of time the markets were more negative, because investors rightly perceived that central banks would not do enough.  So do more!!  In 2007-08 investors were pessimistic because they thought the Fed wasn’t cutting rates fast enough.  The Fed needed faster rate cuts to enable the monetary base to increase.  So do more!! Similarly, in recent months, markets in places like Japan have become pessimistic because the BOJ is not cutting rates fast enough, or is suggesting it may give up. But the media thinks the market is pessimistic because the BOJ is doing negative IOR, even though asset markets respond positively to negative IOR.

In both cases, people wrongly assumed that the problem was that the measures that were taken were not effective.  Instead of, “So do more!” it became perceived as, “So stop doing that, it’s not working.”  They ignored the fact that markets clearly indicated it was working, but that much more needed to be done.  Did I say, “So do more”?

Sometimes I think I’m going crazy—maybe everything I believe is wrong.  After all, almost everything I read is diametrically opposed to what clearly seems to be happening, or to what our textbooks teach us about monetary policy.  But then I look at the asset markets, and am reassured.  I’m not really losing my mind. Monetary stimulus is effective, and it’s needed in Japan and Europe, and was needed in 2007-08 in America.  No matter how many times the press tells us that the markets hate negative IOR, each new IOR news shock confirms once again that the markets prefer even more negative IOR in Europe and Japan.  I don’t have my head in the sand, it’s the business press that does.

Yurt place or mine?

[I am on vacation now, so I dug up an old piece that I never posted.  Also check out the piece on kidney markets, which the WaPo requested me to write.]

Back in 2009 I did a post on “The Aesthetics of Inequality.”  Here’s one excerpt:

A peasant village perched on a hillside in a third world country can be aesthetically beautiful, but a shantytown of former peasants on the edge of a large modern urban area (even if the peasants are now better off) is aesthetically ugly.

I was reminded of this while watching a recent video of Ulaanbaatar, which is attracting mass migration from the Mongolian countryside.  There are few more picturesque scenes of rural poverty that a bunch of yurts set up on Mongolia’s vast plains, under a cerulean blue sky:

Screen Shot 2015-09-21 at 4.01.37 PMFortunately, a mining boom has made Mongolia rich, and the peasants are flocking to the cities, and put up their yurts in shanty towns on the edge of Ulaanbaatar—which now has a majority of Mongolia’s population.

Screen Shot 2015-09-21 at 4.05.00 PM

The lady who narrated the show was a Westerner.  And she couldn’t quite get past the idea that this urbanization didn’t look very attractive:

It’s so funny because in the countryside it seems natural and it seems clean and it seems lovely, and here this is just poverty-struck, and so not natural.

It’s easy for Westerners and/or upper class people to think they know what’s best for the poor. But unless they’ve actually lived that life, they may end up substituting aesthetic judgments for utilitarian criteria. The Mongolians had to decide whether their country was going to be a place of yurts, or a place full of large open pit mines and cities with high-rise apartment buildings.  They chose the latter. The Mongolian she interviewed pointed out that Mongolians wanted to live in these modern apartment buildings because “they don’t have to make fire anymore, they don’t have to carry water anymore.”  They moved to the city to increase their chances of getting a modern place to live.  (Although it looks like most face a long wait.)

The average income of Hispanics is about 79% of the average income of all Americans. The IMF says that per capita income in Spain (PPP) is about 62% of per capita income of the US.  And yet in my mind Hispanics seem “poorer” than Spaniards. My mental image probably reflects the fact that I find Barcelona and Sevilla to be more attractive that the typical Hispanic neighborhood in a large Sunbelt city.  It’s fine to have those images in your mind, as long as your don’t confuse aesthetics with utility.

PS.  My God! Some of my 2009 posts were really long-winded.

PPS.  Yes, my previous example is slightly distorted by the above average size of Hispanic families, but not enough to change my point.

Hindsight is 20-20 (at best)

Update:  I see Matt Yglesias beat me to it.

Before criticizing a Paul Krugman post let me praise his recent post on the Chinese yuan.  I completely agree that the press is overplaying the IMF’s decision to make it a reserve currency.  I did see one report that this might push the Chinese to do more financial reforms, which would be fine.  But countries don’t benefit from reserve currency status anywhere near as much as the media would lead you to believe.

In an earlier post Krugman unintentionally insults Dean Baker:

It’s true that Greenspan and others were busy denying the very possibility of a housing bubble. And it’s also true that anyone suggesting that such a bubble existed was attacked furiously — “You’re only saying that because you hate Bush!” Still, there were a number of economic analysts making the case for a massive bubble. Here’s Dean Baker in 2002. Bill McBride (Calculated Risk) was on the case early and very effectively. I keyed off Baker and McBride, arguing for a bubble in 2004 and making my big statement about the analytics in 2005, that is, if anything a bit earlier than most of the events in the film. I’m still fairly proud of that piece, by the way, because I think I got it very right by emphasizing the importance of breaking apart regional trends.

So the bubble itself was something number crunchers could see without delving into the details of MBS, traveling around Florida, or any of the other drama shown in the film. In fact, I’d say that the housing bubble of the mid-2000s was the most obvious thing I’ve ever seen, and that the refusal of so many people to acknowledge the possibility was a dramatic illustration of motivated reasoning at work.

I hear this claim over and over again, and just don’t understand it.  First let’s consider the Baker claim that 2002 was a bubble.  As the following graph shows, US housing prices are higher than in 2002, even adjusted for inflation.  In nominal terms they are much higher.  So the 2002 bubble claim turned out to be completely incorrect. Krugman needs to stop mentioning Baker’s 2002 prediction, which I’m sure Baker would just as soon forget.

Screen Shot 2015-12-01 at 8.06.57 PM

Of course prices then rose much higher, and are still somewhat lower that the 2006 peak, especially when adjusting for inflation. So I can see how someone might claim that 2006 was a bubble (although I don’t agree, because I don’t think bubbles exist.)

But here’s what I don’t get.  Krugman says it was one of the “most obvious” things he’d ever seen.  That’s really odd.  I looked at the other developed countries that had similar price run-ups, and found 11.  Of those 6 stayed up at “bubble” levels and 5 came back down.  If they still reported New Zealand it would have been 7 to 5 against Krugman. How can you claim something is completely obvious, when there is less than a 50-50 chance you will be correct?  If Krugman were British or Canadian he would have been wrong.

But it gets worse.  Almost no one, not even Krugman, thought we’d have a Great Recession. That makes the bubble claim even more dubious.  Does anyone seriously believe that the utter collapse of the Greek economy has nothing to do with the decline in Greek house prices?  That it’s all about bubbles bursting, with no fundamental factors at all?  That seems to be the claim of the bubble mongers. Even in the US, at least a part of the decline was due to the weak economy.  No, I’m not claiming all of it, but at least a portion.  Look at France, which held up pretty well despite a double-dip recession.  You can clearly see the double dip in the French house prices, so no one can tell me that macroeconomic shocks like bad recessions don’t affect house prices.

In conclusion, even ignoring the elephant in the room–the Great Recession–Krugman’s claim that a bubble was obvious makes no sense.  Most housing markets didn’t collapse after similar run ups.  Most are still up near the peak levels of 2006, even adjusted for inflation (and significantly higher in nominal terms.)  But add in the Great Recession, and the bubble claim becomes far weaker.

Of all the cognitive illusions in economics, bubbles are one of the most seductive. But I expect more from an economist who usually sees through these cognitive illusions, and did a good job showing the fallacy of the claim that reserve currencies status has great benefits to an economy.

PS.  In case you have trouble reading the graphs, the 5 countries with big drops are the US, Greece, Italy, Spain and Ireland.  The US drop is even more surprising when you consider that our post-2006 macro performance is more like the 6 winners.  So I could have claimed it was 6 to 1 against Krugman, if I’d put in a dummy variable for “PIIGS” status.  Does any bubble-monger know why Australian, British, Belgian, Canadian, French, New Zealand, and Swedish house prices are still close to the same lofty levels as 2006, or even higher?  What’s different about the US that made a collapse “inevitable”?

(Paging Kevin Erdmann.)

PS.  I also have a post on Krugman over at Econlog, in case you haven’t gotten your fill here.

When do the Dems believe in trickle-down?

Here’s my hypothesis:  When it comes to microeconomics the Dems are the “stupid party”.  When it comes to monetary policy, and just about all non-economic areas of public policy, the GOP is the “stupid party”.

How could we tell if I’m right about the Dems?  We know that economics is really, really counterintuitive.  It doesn’t seem logical that imports would be good for the economy, or that price gouging in a natural disaster would be good for consumers, or that regulations banning banks from charging fees on ATMS would be bad for bank consumers.  But they are.

So let’s suppose that Dems are like most people; they are not very good at economics. And we also know that they claim to favor the “little guy” and have contempt for “trickle-down economics”, which is the idea that sound economic policies will also benefit people at the bottom.

If my theory is correct, then you’d expect the Dems to favor trickle down policies whenever there were easily discernible “concrete steps” linking the subsidies for big business with the welfare of the common man.

Thus Dems would oppose a cut in the corporate tax rate, unless competition from overseas started to raise fears of jobs losses.  And even then they’d demand that any cuts in the top rate be offset by the closing of loopholes.  And that’s exactly what we observe.

Most importantly, Dems would favor subsidies for big business that seemed likely to directly create jobs, such as the Ex-Im Bank.  And guess what, there is far more support for the Ex-Im Bank (an almost perfect example of crony capitalism) among Dems than among the GOP.  Even when not at the zero bound, and hence not at a time where there might conceivably be a net gain in employment.  Stupid.

Another example is the GSEs, Fannie and Freddie.  These companies have traditionally been strongly supported by the Dems, despite their outrageous business model and obscene profits, because they were seen as helping the common man buy a house.  (As an aside, a portion of the GOP agrees with the Dems, but that’s because big business owns a portion of the GOP, not for ideological reasons. The GOP ideologues tend to oppose crony capitalism.)

What about the vast range of issues where the Dems oppose sound economic policies? My claim is that those are areas where the “concrete steps” helping the average guy are harder to see.  More counterintuitive.

I conclude that the Dems actually do favor trickle-down economics, when they understand it, they simply don’t have the imagination required to see the vast array of areas where deregulation, privatization, and tax reform would help the average guy. They can’t envision anything beyond concrete steps.

The current Ex-Im dispute is the “tell” that lets us see into the mind of Dems, to understand what’s actually motivating their supposed “anti-business” worldview.

Update:  A few additional points, based on some of the comments.  Some seemed to think this post was in some way defending the GOP.  It’s far more critical of the GOP than the Dems.  Take another look.  People are also confused about “trickle down economics”.  AFAIK no one ever advocated trickle down economics.  I assumed everyone knew that.  The title of the post was a joke.  “Trickle down” was a term of derision used against some of the early neoliberal policy reform advocates such as Art Laffer and Jack Kemp, who claimed that deregulation, privatization and cuts in high marginal tax rates would help everyone, including the poor.  Again, no one actually advocated trickle down, it was a term of derision by those who didn’t understand neoliberalism, who could not fathom how conservatives could actually believe that supply-side policies would help the poor.  So they simply imagined that conservatives must believe in “trickle down”, which (AFAIK) no one believes. Why else would they favor tax cuts for the rich?

One criticism I do agree with is that this post is stupid.  All my posts on politics are stupid.  Indeed all articles on politics not written by Scott Alexander are stupid. Talking about politics immediately lowers your IQ by 25 points.  That’s why Tyler rarely writes on politics, he’s too smart to write stupid things.

And yes, the previous paragraph is also stupid, sort of.

Fed meeting today.  Daniel Reeves sent me the following movie posters:

Screen Shot 2015-10-28 at 10.56.15 AMScreen Shot 2015-10-28 at 10.55.58 AM