Steve Randy Waldman, aka Interfluidity, is wicked smart. So it was with mild trepidation that I read Nicholas Weininger’s request:
I’d like to see your take on Interfluidity’s posts about inflation-averse influencers, flights to safety, wealth as a positional good for insurance purposes, and the effects thereof on productivity. Relevant links:
http://www.interfluidity.com/v2/3487.html
https://plus.google.com/112482032780181267192/posts/5ATAa1ckps9
http://www.interfluidity.com/v2/3212.html
Alright, let’s see what I can do.
On inflation-averse influencers, I agree with a lot of the post. I’d quibble with the assumption that our current recession is 100% demand-driven. I favor NGDP targeting, and I think our problems would be less acute if we had more monetary stimulus, but I still think a lot of our problems are structural. NGDP targeting is not going to bring back the 90s.
That quibble aside, I think Waldman is totally right that we have a public choice problem. The way I think about it is that there is a public choice curve.
Superimpose the public choice curve over a dynamic AS/AD diagram, properly situated, and you will see why we are stuck. The public choice curve is meant to bolster the argument that recession is ultimately what the polity chooses, but keep in mind also one encouraging and one discouraging thought. The encouraging one is that having a polity that places a ceiling on the rate of inflation it will tolerate is actually an achievement when compared to the alternative. I lived in Brazil in the early 1980s, and while I was too young to really understand the value of money, I remember dropping a few million cruzeiros on a pack of gum. We might rightly call our current public choice curve inverted because there is an inflation ceiling rather than an inflation floor, which is much more common. Again, this is not to deny that we should have more monetary stimulus now.
The most discouraging part of the public choice curve is the bound on growth. There is a lot that a moderately-enlightened dictator could do to increase the growth rate of the economy from the supply side. I wish that people would get as incensed over the political failure on the supply side as they do on the demand side.
On the other post on the insurance value of wealth, I really disagree, not with the proposition that wealth is a kind of insurance for the wealthy, but that this has a significant real effect on employment. I think Waldman elides differences between the real and financial sector and between long and short runs.
Let’s say that Scrooge McDuck goes to his bank, withdraws his billions in $100 bills and puts them in a vault somewhere, as insurance against some improbable bad event. Aside from this, he spends only enough to keep himself alive. What are the real effects of McDuck’s insurance “purchase”? Putting aside any short-run distortions, all he is doing is leaving more real resources for everyone else to consume. Because the money in his vault is not circulating, it has no real long-run effect, at least not to a first approximation. Waldman needs to find some way to reconcile his argument, at least in the long run, with Steve Landsburg’s point about misers.
If you want to say, “OK, the insurance function of wealth does not distort the real economy in the long run, but it has an effect in the short run,” you still have a lot of work to do. First, Waldman needs to walk back the article I referred to in the first half of this post, in which he argues that depression is a choice. Assuming, arguendo, that the wealthy disrupt the circular flow of payments somehow with their insurance demands, why can’t monetary policy fix that? On top of this, I’m not sure that the comparative statics of anything we might plausibly consider insurance work out: you would expect wealthy people to “buy insurance” during good times and “collect a payout” during bad times. But isn’t the problem the reverse, that we observe hoarding in bad times? Finally, inequality is procyclical, so it’s not as if recessions are caused by sudden outbreaks of inequality.
I’m totally willing to consider the possibility that inequality adversely affects the real economy; indeed, I share the sense that most economists probably have that a highly unequal society likely has something wrong with it. I guess I am just an inequality traditionalist who thinks more in terms of plutocracy, unholy alliances between business and government, and standard public choice. While I’m not persuaded by Interfluidity’s account of the inequality-unemployment channel, I’d be eager to read other economists discussing this question.
A couple of points:
1. About cyclical character of insurance demand. You’re right that the demand for insurance should be higher during good times, but what I think you’re missing is the supply side of the “wealth insurance” market. There are more apparent (though not necessarily actual) opportunities to store wealth reliably during good times than bad. This proposition may depend on how you define good and bad times, but surely comparing now to 7 years ago, the general perception then was that there were plenty of safe investments available, whereas today the perception is that there are few. To the extent that supply in this market is more cyclical than demand, we should expect the price of such insurance to be countercyclical.
2. The distinction between the short-run and the long-run gets blurred when the central bank is setting a ceiling on the inflation rate and the natural interest rate is negative. If the central bank is offering an asset with a guaranteed minimum return of negative 2% but people consistently prefer future expenditure over current expenditure with a subjective discount rate of negative 3%, then you’re never going to get to equilibrium, and there is no long run.
I think you’re right that appropriate monetary policy can, at least in theory, eliminate the problem that Steve describes. (I’m not sure NGDP targeting alone would do it, unless the growth rate of the target path is sufficiently high.) That does lead to other problems, though (asset price instability, as present value becomes very sensitive to estimates of future flows when the discount rate is very low). If more equal distribution has the effect of raising the marginal discount rate, then it solves both problems.
Thanks for the comment, Andy!
I take your point that the price of wealth-as-insurance can be countercyclical due to the perceived scarcity of safe assets during bad times. But in Steve’s model, isn’t it the quantity, not the price, that matters? And to the extent that the price of safe assets is high, it pushes investors into relatively lower-priced, risky assets. If the price of wealth-as-insurance stayed low during a recession, it would make recessions worse, not better, right?
Your argument about negative discount rates is true whether or not Steve’s model accurately represents the world. In the absence of monetary policy we might never end up in the long run if your hypothetical were true. But that doesn’t tell us to what extent Steve’s model is true (not that you’re saying it does, exactly).
Wow, thanks for the prompt reply! I should make more such requests.
Steve’s model seems to ignore the possibility that people can be employed in activities that constitute not the production of consumption goods but the transfer of wealth into the future. This is a simplifying assumption which may be reasonable than it sounds at first. Obviously it is possible to build houses and to invest in improvements to production, but at what point does the risk-adjusted marginal product of such activities become so negative as to be inconsistent with low inflation? I think this is the critical issue that relates both to cyclical changes in the price of wealth-as-insurance and to the short-run/long-run distinction.
If wealth distribution is highly unequal, there will be great demand to transfer wealth into the future, which is to say, demand for “insurance.” When the perceived supply of such insurance is high, the “price” will not be too high, which is to say that the risk-adjusted marginal product of non-monetary forms of wealth transfer will not be too low to be consistent with a low inflation rate. When the perceived supply of such insurance is low, the “price” will be too high, and the marginal product of its non-monetary form too low, to be consistent with a low inflation rate.
You can always fix this problem by having a sufficiently high inflation rate, but “sufficiently high” may be higher than we are willing to accept (and, as noted above, there is the problem of asset price instability). I think NGDP targeting give a better shot at fixing the problem without producing excessive inflation, but it isn’t guaranteed. You might end up with cycles of recession and recovery, because the growth rate of the NGDP target path might not be high enough to be consistent with the equilibrium interest rate given the degree of inequality. As you continue to miss targets, the target growth rate keeps increasing, so it will eventually be high enough, but once you hit the target the cycle starts over.