Arnold Kling  

The Stimulus and Black Swans

Getting to Recovery... Macroeconometrics: The Lost H...

I wrote this for Cato.

The present scenario analysis highlights two black swans. The first one is Depression Averted, under which a stimulus keeps the economy from falling in a downward spiral of layoffs and shutdowns. The other black swan is Catastrophic Collapse, in which a loss of confidence by investors in U.S. government debt leads to a total collapse in the U.S. financial system, with economic activity contracting by 90 percent or more.
The case for or against a fiscal stimulus comes down to the relative importance of these two black swans. A stimulus is worthwhile if the probability of Depression Averted is very high compared to the probability of Catastrophic Collapse. Although neither scenario is likely, I believe that Catastrophic Collapse represents the greater risk.

Read the whole thing. My point is not that I think that a sovereign debt crisis is likely if the U.S. enacts the stimulus. But I think it represents a much worse possible outcome.

The sensible policy would be to enact a small but genuine stimulus. The current stimulus bill is both too small (in terms of the actual stimulus) and too large (in what it does to the U.S. fiscal position). Greg Mankiw's idea of cutting payroll taxes while scheduling future increases in the gasoline tax would be better.

Comments and Sharing

COMMENTS (16 to date)
Corey S. writes:

Arnold- the link is broken.

Arnold Kling writes:

The link works for me

Niccolo writes:

Greg Mankiw's plan is good in the first part, bad in the second.

But it's preferable to this.

Bill Seitz writes:

Doesn't cutting the payroll tax further-muddy the line between the federal General Fund and SocialSecurity/Medicare?

Gary Rogers writes:

I wish I could be as optimistic as you are. Right now the catastrophic collapse of our monetary system is all but inevitable because we have no way to live within our income. It is just a matter of time before the inevitable happens and the stimulus brings that collapse ever closer if it does not trigger it. If it were just debt from a one time event like a past war, we could pay down our debt over time. Unfortunately, we have an irresponsible government that cannot control spending which means deficits into the future forever. That is unsustainable which guarantees a collapse.

Right now I would like to be recommending tax cuts to stimulate the economy. This has proven effective in the past, but unfortunately also created most of the debt we are carrying today. Since we never were responsible enough to build our savings back up in the good times the debt has continued to build. After all, a stimulus is nothing more than artificially encouraging businesses and consumers to spend over saving. If you stimulate every time there is a downturn but never encourage saving in the up cycles, you end up in debt with no savings. That is where we are today and even though the catastrophic collapse has not happened yet, any increase in debt moves that day closer.

I think the black swan in this case is having our legislators start showing leadership and demonstrate that we can live within our means. The probability of it happening is slim, but it is a possibility. And, on the positive side, if we do that and quit borrowing dollars back from overseas, those dollars would be available to buy American goods, which means American jobs. Isn't that the goal?

jb writes:

Gary, what's this 'we have no way to live within our income' BS? I have lived within my income just fine for quite a while, and many of my friends who were "barely in the red" (slowly accumulating a small amount of debt) are forcing themselves back into solvency by cutting back.

I see 18 cars in line at the "custom hand wash" car washes around here. anecdotal, but reassuring to me. Home sales are up. Spending on luxury goods are way down. Companies are slashing prices, even on new goods. Demand for gas continues to stagnate. People are saving money, and the reductions in prices are helping.

There's no doubt that the leadership is completely lacking. But what I'm starting to wonder is - how many americans are already so contemptuous of Washington that they don't care either way about the stimulus? That they will live responsibly and use their own judgement to decide when to spend and when to save, instead of Washingtons?

And, if enough people are angry enough about the deficits, we'll see some progress on taming them. That's a big if, I know, but it is less of a "big if" than it was 5 years ago.

Gary Rogers writes:

I had no intention of accusing anyone other than the government of mismanaging their finances. Some do better than others, but we all do the best we can to make good decision and end up living with our choices. But, if you think that you are not going to be held responsible for the $10 trillion of debt your representatives in Washington have piled up, you are being misled. Government debt is real debt and must someday be repaid through higher taxes or selling assets.

fundamentalist writes:

Dr. Kling assumes that the stimulus will either avert the depression or not. To insert the Austrian view on the swans, he needs a category that says the stimulus will make the depression worse. But according to Austrian econ, that is not a black swan, but a huge white one.

He adds the remote possibility that the huge debt will cause people to lose confidence in the US dollar and federal debt. That is more probable than a black swan. The most likely outcome will be that the stimulus makes the depression go deeper and last longer than it would have without the stimulus, and then people will lose confidence in federal debt because of its size.

Lee Kelly writes:
A stimulus is worthwhile if the probability of Depression Averted is very high compared to the probability of Catastrophic Collapse. - Arnold Kling

Even if the probability of a fiscal stimulus averting a depression were very high compared to the probability of a catastrophic collapse, it does not follow that a fiscal stimulus is worthwhile.

The fiscal stimulus is only worthwhile if it actually does avert a depression. Suppose that the probability of it doing so were high compared to the probability of a catastrophic collapse. But since improbable events are possible events, further suppose it does not avert a depression, and in fact, contributes toward a catastrophic collapse.

Would you actually turn around as say, 'oh well, it was worth it'? Of course not, because it is only worth anything if it actually works. Probability has nothing to do with it, but only with what some people anticipate will work. But a wrong decision is a wrong decision even when some subjective probability evaluation deems it favourable.

aaron writes:

A gas tax will be a bad idea until some inefficiencies are addressed.

Fuel economy delined, despite less driving and more efficient cars being purchased, since gas prices started sky-rocketing in '06.

Some reasons are:

Change in fleet, efficient cars of early-mid 90's may be retiring. SUVs of the late 90s and early 2000 may make up a greater portion of the fleet despite declining sales.

Cell phone usage. Cellphone users increase following distances and are slower to get up to speed. This decreases the capacity of roads and decreases throughput at bottlenecks, leading to more stop-and-go traffic and congestion.

But I think the main cause has been bad reation to high gas prices. People mistake slow for efficient.

Accelerating quickly, but smoothly, is generally more efficient than slow. Higher cruising speeds are more efficient below 55mph, provided the cruising time is long enough to make up for the cost of the additional acceleration. This is contrary to popular belief. And slow driving has the same effects on traffic that make cell phone usage bad for fuel economy.

Giffen Behavior, people driving more during congested times to maintain their income, and driving less for consumption during times of excess capacity.

Then there are instances where good behavior becomes bad. Avoiding brakes is a very economical behavior, provided there is no congestion. But idling along because of expecting a to hit a traffic signal a second time prevents other queues from clearing and prevents people from reaching destinations that might be before the bottleneck.

This is very important considering the negetive effects of congestion on economic growth.

MHodak writes:

I think the biggest problem with this debate, and much of what passes for policy making, is the lack of analysis of these alternatives.

I think Kevin Murphy did an outstanding job of proposing a decent, if incomplete, analytical framework for thinking about the stimulus, down to a plausible equation, and assumptions about its terms. Various thinkers (Murphy, DeLong, etc.) disagreed about the terms, but at least it got us thinking like economists about economic consequences, instead of just a bunch of people with varying opinions.

No one knows, of course, what the chances of recovery or collapse are, but there is a lot of data out there. We have index prices, Treasury CDS, etc. that can inform those issues much better than "I think." I wish some economist would take the time to develop a conversation around such a framework. If you know of one, please let me know.

Gongability writes:

Gary Rogers made the following false statement:

"if you think that you are not going to be held responsible for the $10 trillion of debt your representatives in Washington have piled up, you are being misled. Government debt is real debt and must someday be repaid through higher taxes or selling assets."

I used to live in California, and during that time the government accumulated debt, and I defaulted on my portion of that debt when I moved to another state; and a new migrant to California accepted my share. Some time in the coming years, I will emigrate to another country and default on my portion of the federal debt.

The U.S.A. will collapse if and when there are sufficient people leaving without replacements.

Dog of Justice writes:

I'm far from a fan of the stimulus as currently constituted, but wow, it's not every day that I see an argument as absolutely wrong as Lee Kelly's.

Matt C writes:

I am surprised that you see a loss of confidence in the dollar as unlikely in the long term. To me it is quite likely. Almost certain, if we include the (optimistic) scenario where confidence declines slowly rather than suddenly.

We (in the U.S.) have large amounts of public debt, huge amounts of private debt, and an entitlements crisis looming. We won't address any of this unless it is forced on us, which will probably be too late for investor confidence. The only thing you can say in our favor is that most of the rest of the Western world is about as bad. I don't think that will save us, though we might learn something from watching the U.K. go first.

On the other hand, I think you overstate the catastrophic consequences of a sharp loss in dollar credibility. We have seen sovereign debt crises before and they are ugly, but not "90% decline in economic activity" ugly. The U.S. is bigger than Korea or Argentina, but that has pluses as well as minuses. I think (hope) you are too pessimistic here.

I'd be interested in reading a post on why you think the U.S. government will continue to have AAA credit into the indefinite future.

Or, if it is more interesting to you, why a major decline in the credit of the U.S. would be many times worse than the Great Depression, rather than something more comparable to what we're going through right now.

Thanks for veering toward a topic of particular interest to me. :)

Grant Gould writes:

These "black swan" things seem very convenient and in demand these days, at least among rich Washington types. Perhaps I should look into investing in some sort of black swan factory...

Tom writes:

I don't think you understand the concept of 'black swans.' You can't list off two potential black swans because the term 'black swan' refers to unknown unknowns-- situations where you don't know the model, the probabilities or the potential outcomes. A black swan is something nobody expects.

You're talking about two extreme outcomes (nothing bad versus everything bad). Both extremes are widely discussed, so I'm not sure how they fit the definition.

From a review Taleb links to on his website (he refers to it as a "Great & honest bad review ,by a smart but angry reviewer" ):
"But this isn't a treatise on the art of the possible in war. The Black Swans of the title aren't simply known unknowns; [they] are unknown unknowns - events, or inventions, or runaway successes, or indeed contingencies of any kind - for which no statistical analysis, or inductive reasoning can possibly arm us. They are events like 9/11, or Black Monday, or publishing phenomena like the Harry Potter books, or inventions such as the internet, all of which alter the human world."

Comments for this entry have been closed
Return to top