Scott Sumner  

Economics: Truer than it seems

PRINT
Boudreaux on Piketty... My Life of Appeasement...

One thing that economists and non-economists have in common is that they underestimate the power of economic theory. More specifically, they aren't able to accurately connect the predictions of economic theory with the behavior that they (wrongly) think they observe in the people around them. There are many examples of this. For instance, most people (economists and non-economists) underestimate elasticities. They underestimate how strongly people respond to changing incentives, such as tax changes, wage changes, price changes or interest rate changes.

Economists and non-economists also underestimate just how rational people really are, at least in aggregate. How much they understand about the world. And how efficiently markets aggregate information. And they do so because people don't seem very smart, or very rational. As always in economics, appearances can be deceiving.

A good example is Ricardian equivalence (RE.) I thought of this when reading an excellent post by David Glasner. In the comment section of the post a commenter who identified him or herself as a non-economist was dismissive of the whole idea (that government borrowing leads people to save more, in anticipation of the tax burden of future deficits.) This particular individual didn't really understand the subject, but even economists who do understand RE tend to be dismissive.

I'm agnostic on the validity of RE, and certainly don't wish to defend it in this post. Instead, here I'd like to focus on the fact that people dismiss the idea for the wrong reason. They don't find it plausible that people would save more if the budget deficit increased, and indeed they are even skeptical that the public would be aware of the fact that the deficit increased.

If I ask my students whether they find RE to be plausible, they almost all answer "no."

But suppose I ask them a different set of questions:

1. Do you expect to get back all the Social Security that you have been promised?

Almost all say they expect less than what has been promised.

2. Then I ask whether that perception affects their willingness to save money for retirement, outside of Social Security.

Almost all say they are more likely to save for their retirement because of the perception that Social Security is on the road to bankruptcy.

But if you put these two answers together then you have Ricardian Equivalence! To see why, let's first ask where the perception of Social Security bankruptcy is coming from. Most likely it is the media, which harps on the debt crisis to the point where Paul Krugman wants to pull his hair out. (BTW, like Krugman, I believe the debt problem is less severe than the Very Serious People in the media suggest.) But what Krugman and I believe is not important. What is important is that the American public is well aware of the US debt problem, and also aware than many experts think that it will eventually lead to cuts in Social Security. And that perception does lead Americans to save more, just in case.

Do American save more or less than the extra amount predicted by Ricardian Equivalence? There is absolutely no way to answer that question, because it involves not one but two great unknowns:

1. How bad is the debt situation? (No one knows, not even Larry Kotlikoff.)
2. How much does worry about the debt impact American saving?

I believe most Americans overestimate the problems with Social Security. That suggests a sort of super RE, even more saving than predicted. But I also believe Americans might save less extra money than would be rational, based solely on the prediction of RE. There are other factors such as borrowing constraints. And then there are marginal effects to consider---if the debt problem gets worse, how do Americans respond at the margin to increasingly hysterical media portrayals of the debt situation.

We can't answer these questions. In that case, we have no reason to assume non-RE, and rationality is probably the best baseline assumption to work with, unless we have empirical evidence suggesting that deficits raise interest rates. And most of the empirical work I saw when this was a hot issue suggested that deficits don't impact interest rates, just as RE predicts.

What's true of RE is also true of efficient markets, rational expectations, the supply-side effects of high MTRs and lots of other areas of economics. People respond more strongly to incentives than you'd think. They are more rational than you'd think.

Common sense is almost useless in the field of economics. Never dismiss an economic theory because the assumptions about human behavior don't sound plausible.

After all, quantum mechanics doesn't sound plausible either---but it's true.


Comments and Sharing





COMMENTS (12 to date)
Kevin Erdmann writes:

People say that the housing market in the 2000's was a bubble because they knew a guy who was the night janitor at the local YMCA who was flipping houses during the day because he said, "Real estate never goes down." Those people have apparently never seen the Yahoo Finance message boards.

Evolution created us without any rational forethought whatsoever. In like fashion, markets can find efficient equilibria as long as some portion of human agents can at least act randomly. Though, I admit this is not a sure bet.

Scott Adams had a funny post on this the other day:
http://dilbert.com/blog/entry/the_illusion_of_intelligence/

Also, on RE, if we come at it from the other direction, isn't it the obvious starting point? After all, all else equal, when the government takes on debt, a T-bill is created in that exact amount. As a first order effect, doesn't it have to be a 1-1 RE? So, isn't RE the wrong question to ask? Isn't the right question simply to ask what the demand for treasury debt is, and whether that leads to producer surplus for the government? For instance, by providing low risk savings vehicles for developing world capital? So, it basically comes down to, what's the price?

John Thacker writes:

I don't think that Social Security is an insurmountable problem, but it's worth noting that all those projections that say that Social Security become sustainable assume a sudden ~20% cut in Social Security payments when the Trust Fund is exhausted. So I think it's entirely consistent to say that "Social Security isn't as big a deal as the media claims / the problem is manageable" but also "I don't expect to get all the promised Social Security."

Scott Sumner writes:

Kevin and John, Good points.

Kevin, I see your point as part of a very complex argument over RE. It's one of many possible factors that could make RE less than perfect.

John, FWIW, my best guess is that the lower income people will get essentially all they were promised, and that the program will be fixed with a combination of higher taxes and modest cuts to benefits for the rich.

Kevin Erdmann writes:

Well, I'm mostly trying to agree with you. It seems like the best understanding is to start at RE and look for deviations, as opposed to assuming it doesn't hold from the get go.

Daniel Kuehn writes:

I agree generally with this, but economic theory also seems to show that a little bit of sand in the gears can have substantial consequences. For that reason I think skepticism about perfect Ricardian Equivalence despite a fair amount of faith in generally rational expectations is a meaningful distinction that I bet a lot of economists maintain.

I know "a little bit of sand in the gears" is an amazingly precise statement on my part, but I think you get my point :)

Kevin Erdmann writes:

I don't get these discussions about rational expectations or a binary idea of whether government borrowing is positive or negative. If government spends money via borrowing, as a first effect $ will be drawn into t-bills from somewhere else. No rational expectations required.
Now, assuming some sort of systematic behavior means that savers will try to achieve the same level of risk that they had before the t-bills entered the market, so they will add some leverage to retain some amount of the previous private investment.
there will be some marginal effect on interest rates to the extent that this leveraging happens and there is some value to the spending compared to whatever private spending happened before. So it comes down to price. And it takes rational expectations to get to a place where Ricardian Equivalence might not hold.
Being skeptical of RE because one is skeptical of rational expectations is like being skeptical of IBM stock without knowing the price.

Joe Teicher writes:

If your students are like normal college students they haven't worked long enough to be promised anything by social security. Perhaps non-economists doubt the value of economic theory because they see how little economists sweat the details.

Michael Byrnes writes:

Excellent post.

Before reading this I would have said I was skeptical about rational expectations. But after reading this post and your earlier one on the same topic (link: http://econlog.econlib.org/archives/2014/05/rational_expect.html ) I think I am starting to understand your argument.

"Rational expectations" doesn't rely on the brilliance and foresight of all of the individuals in a market... it just relies on the market itself (and the incentives it provides), which will tend to arrive at the right answer even if the most of the individuals involved do not even know the question, even if many act irrationally, even if it would not have been possible for anyone to predict that answer beforehand.

It makes great sense, although I also think Daniel Kuehn is right about sand in the gears.

Brent writes:

In the traditional textbook model, isn't it the case that budget deficits are treated as saving? Thus, when deficits increase, at least the domestically financed portion is causing increased saving... no?

Scott Sumner writes:

Daniel, My problem isn't skepticism about RE (I'm skeptical too) but rather jumping to the conclusion that deviations from RE are in one direction rather than another.

Kevin, The easiest way to think about it is that opponents of RE think the public treats government bonds as net wealth, and that causes the public to attempt to save less.

Joe, Not sure if you are joking.

Thanks Michael.

Brent, Deficits are treated as dissaving.

Steve writes:

RE works for me in a much simpler way. I save/invest all excess income after paying my fixed expenses. If the deficit increases via lower taxes, then my excess income increases and so in turn does my savings.

Jason Smith writes:

Is it that we are more rational than we think, or is it that we are irrational but we are pushed by rational market forces that are independent of human decision-making?

http://informationtransfereconomics.blogspot.com/2014/08/rationality-and-entropic-forces.html

Comments for this entry have been closed
Return to top