Murphy's on a roll on interest rates

A very good post from Bob Murphy continuing the assault on Wenzel. I wanted to highlight this point specifically:

"The standard (and obvious) explanation for the yield curve invokes the desire for liquidity. I’m guessing Wenzel will come back and say something like, “Short rates might move in the future, so that’s why the yield curve can be upward or downward sloping.” But my point is, even if we expected the short rate to stay constant for the next ten years, a desire for liquidity would cause the yield curve to be upward sloping. If you roll your money over 10 times, you are less exposed to a sudden (and unexpected) move in interest rates than if your money is “stuck” in a ten-year bond. If for some reason your plans change, and you need to spend your money before the originally planned ten years, and if interest rates have risen in the meantime, you will take less of a hit if you have been rolling your money over in one-year bonds, than if it’s still sitting in (say) a 60%-matured 10-year bond."

This point that Bob makes is precisely the connection between uncertainty about the future and liquidity preference that Keynes makes, and I thought that was worth pointing out. In chapter 13 Keynes writes:

"At this point, however, let us turn back and consider why such a thing as liquidity-preference exists. In this connection we can usefully employ the ancient distinction between the use of money for the transaction of current business and its use as a store of wealth. As regards the first of these two uses, it is obvious that up to a point it is worth while to sacrifice a certain amount of interest for the convenience of liquidity. But, given that the rate of interest is never negative, why should anyone prefer to hold his wealth in a form which yields little or no interest to holding it in a form which yields interest (assuming, of course, at this stage, that the risk of default is the same in respect of a bank balance as of a bond)? A full explanation is complex and must wait for Chapter 15. There is, however, a necessary condition failing which the existence of a liquidity-preference for money as a means of holding wealth could not exist.

This necessary condition is the existence of uncertainty as to the future of the rate of interest, i.e. as to the complex of rates of interest for varying maturities which will rule at future dates. For if the rates of interest ruling at all future times could be foreseen with certainty, all future rates of interest could be inferred from the present rates of interest for debts of different maturities, which would be adjusted to the knowledge of the future rates. For example, if 1dr is the value in the present year 1 of £1 deferred r years and it is known that ndr will be the value in the year n of £1 deferred r years from that date, we have

ndr = ndn+r / 1dn ;

whence it follows that the rate at which any debt can be turned into cash n years hence is given by two out of the complex of current rates of interest. If the current rate of interest is positive for debts of every maturity, it must always be more advantageous to purchase a debt than to hold cash as a store of wealth.

If, on the contrary, the future rate of interest is uncertain we cannot safely infer that ndr will prove to be equal to 1dn+r/1dn when the time comes. Thus if a need for liquid cash may conceivably arise before the expiry of n years, there is a risk of a loss being incurred in purchasing a long-term debt and subsequently turning it into cash, as compared with holding cash. The actuarial profit or mathematical expectation of gain calculated in accordance with the existing probabilities — if it can be so calculated, which is doubtful — must be sufficient to compensate for the risk of disappointment.

There is, moreover, a further ground for liquidity-preference which results from the existence of uncertainty as to the future of the rate of interest, provided that there is an organised market for dealing in debts. For different people will estimate the prospects differently and anyone who differs from the predominant opinion as expressed in market quotations may have a good reason for keeping liquid resources in order to profit, if he is right, from its turning out in due course that the 1dr’s were in a mistaken relationship to one another.

This is closely analogous to what we have already discussed at some length in connection with the marginal efficiency of capital. Just as we found that the marginal efficiency of capital is fixed, not by the “best” opinion, but by the market valuation as determined by mass psychology, so also expectations as to the future of the rate of interest as fixed by mass psychology have their reactions on liquidity-preference; — but with this addition that the individual, who believes that future rates of interest will be above the rates assumed by the market, has a reason for keeping actual liquid cash, whilst the individual who differs from the market in the other direction will have a motive for borrowing money for short periods in order to purchase debts of longer term. The market price will be fixed at the point at which the sales of the “bears” and the purchases of the “bulls” are balanced."

This passage, which actually comes right after the Chapter 13 passage that Bob quoted earlier, was one of those lightbulb moments for me when I first read the book. OK, now I'm convinced that this fall I need to:

1. Reread the General Theory, and
2. Read Bob's dissertation

Bob Wenzel on Bob Murphy on Liquidity Preference

Bob Wenzel criticizes Bob Murphy for drawing an objective conclusion regardless of the ideological fallout that it might inspire (which is absolutely not to say that everyone who took issue with Bob Murphy was being ideological about it... but you could sense him bracing himself in the original post).

I left this response on Wenzel's blog - nothing revolutionary, but hopefully helpful:

"You seem to be confusing "liquidity" with "cash in your wallet". Liquidity for Keynes is simply the control over funds to use as a medium of exchange.

You can save $1,000 in your mattress for ten years, you can save $1,000 in a savings account for ten years, you can save $1,000 in a 2 year CD for ten years (rolling it over four times), you can save $1,000 in a 5 year CD for ten years (rolling it over once), or you can save $1,000 in a 10 year CD for ten years.

In each case your time preference is to forgo $1,000 in the present and use it ten years from now. However, your liquidity preference varies in each case.

Holding time preference constant, we still have a variance in the interest rate that people are willing to accept.

That's the liquidity preference theory of interest. That's all it is. It shouldn't be all that controversial. People should be able to identify this in their own lives - think about how much you keep in a savings account vs. how much you tie up in other accounts. Usually it's not because the amount you keep in savings is expected to be used any more imminently than what you keep in a CD or something else. Sometimes people have a specific imminent purchase in mind, but often it's intended to sit and earn interest for the same amount of time. You just trade off the interest rate against how accessible you want those funds. If you need less accessibility you tie it up in a higher interest rate account for the exact same time span that you're expecting to leave the money in savings. It's precisely the uncertainty of those expectations that causes us to keep money liquid. If we were certain, we'd put it all in a high interest account for precisely the time period that is consistent with our time preference."

One of the things that bothers me about the way people talk about liquidity preference is when they refer to it as "hoarding". Hoarding implies that mattress-stuffing caused the Great Depression. This is silly, of course. Certain sorts of money holding can be more liquid than others, and so an increase in liquidity preference can shift the composition of money holdings to be relatively more liquid across the board without a discernable increase in cash holdings.

I think it's also fair to give Bob Murphy a voice here. This is his comment on Wenzel's post:

"I was going to make a Darth Vader joke, but this is actually an important issue and I'm amused at how many people are high-fiving Wenzel here, when he is the one who is clearly using a weird, non-layman's definition of "saving."

15-year-old Johnny mows my lawn every week, and I pay him $20 each time. Every week, he spends $15 of it going to the movies with his friends, but he puts $5 in a piggy jar on his bureau.

After a year, he has accumulated $5x52 = $260 which he uses to buy a nice watch. Johnny says, "I'm sure glad I consumed less than my income all year, saving $5 per week. Then I used my accumulated savings to buy a watch. I deferred consumption all year in order to buy a nice good later on."

Wenzel says, "What the heck are you talking about? Are you a Keynesian Johnny? You haven't saved at all."

Are you guys all comfortable with that? You don't think Johnny was saving $5 per week?"

Some clarification on "liquidity traps"

Gary confuses liquidity preference theory and liquidity traps in the comment section of this post. I want to clear up exactly what we're talking about here, and why liquidity preference is very important (whereas liquidity traps are just an interesting curiosity).

Gary writes: "Daniel, While liquidity preference theory is not controversial amongst a subset of financial economists, how often it happens is. This was Milton Friedman's point; yeah, it is possible to get into a liquidity trap (though it is not very likely to ever occur), but to build your entire notion of economics around it (as Keynesians of all stripes do) is silly."

Liquidity preference theory says that one determinant of interest rates is the tradeoff that people make between holding and parting with liquidity. Keynes thought this was the determinant of interest rates. Most people since Keynes (myself included) think it is a determinant of interest rates, but not the only one. If liquidity preference is the only determinant of interest rate, there's no guarantee that the loanable funds market will clear or be consistent with full employment. If interest rates are determined jointly with the market for loanable funds, then the loanable funds market will clear but there's still no guarantee that it will be consistent with full employment.

That's the liquidity preference theory of the interest rate, but that has nothing at all to do with liquidity traps. You could firmly believe liquidity traps are impossible and still have a liquidity preference theory of the interest rate and a Keynesian perspective on the macroeconomy.

So then what's a liquidity trap? A liquidity trap is a situation where the demand for liquidity or money is virtually unlimited - where the money demand curve is highly elastic. Liquidity is in such high demand that no matter how much liquidity is pumped into the market, interest rates don't lower. The demand curve is not downward sloping, it is horizontal. People can't be satiated.

This is the traditional liquidity trap, at least. This is the version that Keynes speculated on briefly. Today, though, the liquidity trap has become associated with the "zero lower bound" thanks to Paul Krugman. In the 1990s, Krugman pointed out that in Japan a nominal interest rate floor acted like a horizontal money demand curve. You can think of the zero lower bound as a price floor in the market for liquidity. Money demand may not be horizontal at all, but the zero lower bound makes it look like we're in a region where it is. I've outline both of these situations (the traditional case and the Krugman case) below in red:



Most Keynesians are pretty skeptical of liquidity traps. We only think two have happened in U.S. history, with a few other cases in other countries. Here's the thing, though: all of these cases have been of the Krugman/ZIRP/faux-liquidity-trap variety. Keynes's original skepticism about a horizontal money demand curve seems to have held up pretty well. We're at the point now where the Krugman version is essentially taken to be synonymous with "liquidity trap" because it's the only version we ever see!

Liquidity traps are very bad news. I'm not of the opinion that they make monetary policy impossible. You can always create inflation with monetary policy, which lowers real interest rates even if nominal interest rates are stuck at zero. You can also do unconventional stuff like charge negative interest rates on reserves held at the Fed. However, it does make a much stronger case for fiscal policy. Creating inflation is hard in the current depressionary environment, so while in theory it's plausible to lower real interest rates that way it's a dicey proposition. Why mess with this horizontal money demand curve when we can increase public investments easily and when there are lots of worthwhile public investments to make? Why mess with a stubborn LM curve when the IS curve is easy enough to shift? So the liquidity trap/ZIRP situation is relevant to Keynesians insofar as it tips the scales in favor of fiscal policy. It has nothing to do with the broader Keynesian point about liquidity preference theory and the cause of the recessions.

A good source on this is Boinavosky (2004).

It's interesting - Krugman has been celebrated for his trade theory, but in the end he will probably be most remembered for permanently switching us from the old view of liquidity traps to the ZIRP view - first as it applied to Japan, and now as it applies to the U.S..

Nick Rowe on Liquidity Data

Nick Rowe has recently climbed several slots in my list of favorite bloggers with a lot of great posts recently. This one on the need for considering liquidity data ("L-data") is especially good and important:

"Economists have got lots of P-data and Q-data, and we pay a lot of attention to it. I think we don't have much L-data, except anecdotal, and we don't pay much attention to the hard L-data we do have.

P-data is data on the prices at which goods are traded. Q-data is data on the quantities of goods traded. We've got it, and we use it. And it's good we've got it and we are right to use it. P-data and Q-data are important. But they are not the only data that are important.

What do I mean by L-data? I'm going to come at that slowly.

Think about the "stylised facts" of the business cycle. In a recession, output and employment fall, or rise less quickly than before. That's Q-data. Prices and wages also fall, or rise less quickly than before. That's P-data. Looking at the P-data and Q-data can help us test theories and understand the causes of the business cycle. But there's some other data that isn't P-data or Q-data that has a big impact on why I think about business cycles the way I do. We ought to be able to explain why we believe what we do believe, and I can't fully explain why I believe that business clcyles are largely demand-driven without talking about L-data.

When we go into a recession, many things become easier to buy and harder to sell. And when we go into a boom, those same things become easier to sell and harder to buy. A recession has lots of buyers' markets and a boom has lots of sellers' markets. That's what I mean by L-data. There's something more going on than what is captured in the P-data and Q-data. There's something more going on than the P-data and Q-data that tell us all we need to know about perfectly competitive markets for perfectly liquid goods with perfectly flexible prices. And that something more is crucial to the way I think about the business cycle....

I think that prices and wages are sticky. And I think that sticky prices and wages are important in understanding the business cyle. Those two things go together. The main reason I think that prices and wages are sticky is not just that they look sticky, but because if I assume that prices and wages are sticky i can make sense of the fact that we get buyers' markets for goods and labour in a recession, and sellers' markets for goods and labour in a boom. If aggregate demand falls, either prices and wages fall, or we get buyers' markets for goods and labour, or we get a bit of both. If aggregate demand rises, either prices and wages rise, or we get sellers' markets for goods and labour, or we get a bit of both. And we generally get a bit of both, in both recessions and booms, though the proportions vary from market to market. And because of imperfect competition, with sellers usually having market power to set prices on average above competitive equilibrium, buyers' markets are normally more common, on average over the business cycle, than sellers' markets."

He goes on to ask for examples of liquidity data that are out there. Several financial examples are given, which isn't surprising - bid-ask spreads, etc. I think the components of the Beveridge Curve - some measure of job seekers, matches, and vacancies is going to be important. There are good points in the comment section - it's worth reading through.

Carl Menger: Keynesian?

We'll never know. He died in 1921. But a commenter on a fantastic post by Nick Rowe (see my next blog post) points out that Menger was very tapped into the idea of liquidity preference:

"But the fact that different goods cannot be exchanged for each other with equal facility was given only scant attention until now. Yet the obvious differences in the marketability of commodities is a phenomenon of such far-reaching practical importance, the success of the economic activity of producers and merchants depending to a very great extent on a correct understanding of the influences here operative, that science cannot, in the long run, avoid an exact investigation of its nature and causes."

That screams "effective demand" of course. Would it have lead him all the way to a liquidity preference theory of the interest rate? If he didn't find his way there himself, it certainly seems like he would have liked a lot of what Keynes had to say.

Jonathan Catalan on the Liquidity Trap

Jonathan's long awaited article on Krugman and the liquidity trap is now up at mises.org. Rather than skim it furiously now, I'm going to wait to read it more carefully and then perhaps share any thoughts I have. I'm seeing lots of familiar citations in the list of references, which is good! The liquidity trap has a long and circuitous theoretical history that is only complicated by the fact that we haven't seen many of them in practice. I imagine there are more theoretical versions of the liquidity trap than there are historical instances of it, in fact! A lot of people gloss over this and provide inappropriate analysis of the problem in the process - it looks like Jonathan is avoiding that.

I like his first footnote especially: "While Paul Krugman is a Keynesian, not all Keynesians agree with Paul Krugman. As such, any Keynesian reader who takes offense at the criticism aimed at Krugman and equivocated with general Keynesian theory should recognize that this characterization is meant for the sake of simplicity."

I have to wonder - was he thinking of me? Anticipating criticism, clarifying, and qualifying is always very good practice. In other words: always make sure you cover your ass.

Indiviglio on Uncertainty

David Indiviglio, at The Atlantic, shares an interesting looking paper arguing that recessions cause uncertainty and not vice versa. What does this mean for a Keynesian? I have a few thoughts:

1. Well it could mean the Keynesian emphasis on uncertainty about the futureis wrong. I don't think it means that, and even if this one paper means that I don't think one paper overthrows an enduring theoretical framework with a lot of explanatory power, but let's start by being up front and making clear (if it was unclear to any readers) that this presents a challenge.

2. I'd have to read the paper, but this could just be a better explanation of the way declining animal spirits work. "Animal spirits" is a vague and nebulous term. I don't think there's anything wrong with saying that collapse in asset values or a real downturn or a downturn in consumer demand causes a decline in animal spirits and an increase in uncertainty which worsens a downturn. This whole "secondary depression"/"feedback loop" mechanism is very common across a lot of business cycle theories, including Keynesian ones.

3. The Keynesian understanding of uncertainty is primarily tied to liquidity preference, and the idea is that even in good times liquidity preference keeps us below our output potential. I'm not sure how any paper could pick up this "baseline" liquidity preference, which is really the primary way that it comes into the Keynesian model.

4. I wonder how macroeconomic policymaking plays into the paper. In the United States, we actually have taken macroeconomic stability more seriously as a policy objective than a lot of the rest of the world (see the Crooked Timber article I shared earlier on Keynesianism as a substitute for social democracy). If macroeconomic policy is done right and it's not taken into account in the model then I'm sure the effect of uncertainty would be weakened.

5. The paper only looked at the manufacturing sector. Not sure what the implications of this are.

Real Time Economics also picked up the paper, and had this to say:

"One conclusion from the paper is that policy makers can talk about the need to end uncertainty all they want, but jaw-boning won't make much difference. Only increased demand will make business executives feel more confident."

Indiviglio also writes on consumer confidence here.