Archive for the 'Scott Sumner' Category

Monetarism and the Great Depression

Last Friday, Scott Sumner posted a diatribe against the IS-LM triggered by a set of slides by Chris Foote of Harvard and the Boston Fed explaining how the effects of monetary policy can be analyzed using the IS-LM framework. What really annoys Scott is the following slide in which Foote compares the “spending (aka Keynesian) hypothesis” and the “money (aka Monetarist) hypothesis” as explanations for the Great Depression. I am also annoyed; whether more annoyed or less annoyed than Scott I can’t say, interpersonal comparisons of annoyance, like interpersonal comparisons of utility, being beyond the ken of economists. But our reasons for annoyance are a little different, so let me try to explore those reasons. But first, let’s look briefly at the source of our common annoyance.

foote_81The “spending hypothesis” attributes the Great Depression to a sudden collapse of spending which, in turn, is attributed to a collapse of consumer confidence resulting from the 1929 stock-market crash and a collapse of investment spending occasioned by a collapse of business confidence. The cause of the collapse in consumer and business confidence is not really specified, but somehow it has to do with the unstable economic and financial situation that characterized the developed world in the wake of World War I. In addition there was, at least according to some accounts, a perverse fiscal response: cuts in government spending and increases in taxes to keep the budget in balance. The latter notion that fiscal policy was contractionary evokes a contemptuous response from Scott, more or less justified, because nominal government spending actually rose in 1930 and 1931 and spending in real terms continued to rise in 1932. But the key point is that government spending in those days was too meager to have made much difference; the spending hypothesis rises or falls on the notion that the trigger for the Great Depression was an autonomous collapse in private spending.

But what really gets Scott all bent out of shape is Foote’s commentary on the “money hypothesis.” In his first bullet point, Foote refers to the 25% decline in M1 between 1929 and 1933, suggesting that monetary policy was really, really tight, but in the next bullet point, Foote points out that if monetary policy was tight, implying a leftward shift in the LM curve, interest rates should have risen. Instead they fell. Moreover, Foote points out that, inasmuch as the price level fell by more than 25% between 1929 and 1933, the real value of the money supply actually increased, so it’s not even clear that there was a leftward shift in the LM curve. You can just feel Scott’s blood boiling:

What interests me is the suggestion that the “money hypothesis” is contradicted by various stylized facts. Interest rates fell.  The real quantity of money rose.  In fact, these two stylized facts are exactly what you’d expect from tight money.  The fact that they seem to contradict the tight money hypothesis does not reflect poorly on the tight money hypothesis, but rather the IS-LM model that says tight money leads to a smaller level of real cash balances and a higher level of interest rates.

To see the absurdity of IS-LM, just consider a monetary policy shock that no one could question—hyperinflation.  Wheelbarrows full of billion mark currency notes. Can we all agree that that would be “easy money?”  Good.  We also know that hyperinflation leads to extremely high interest rates and extremely low real cash balances, just the opposite of the prediction of the IS-LM model.  In contrast, Milton Friedman would tell you that really tight money leads to low interest rates and large real cash balances, exactly what we do see.

Scott is totally right, of course, to point out that the fall in interest rates and the increase in the real quantity of money do not contradict the “money hypothesis.” However, he is also being selective and unfair in making that criticism, because, in two slides following almost immediately after the one to which Scott takes such offense, Foote actually explains that the simple IS-LM analysis presented in the previous slide requires modification to take into account expected deflation, because the demand for money depends on the nominal rate of interest while the amount of investment spending depends on the real rate of interest, and shows how to do the modification. Here are the slides:

foote_83

foote_84Thus, expected deflation raises the real rate of interest thereby shifting the IS curve to the left while leaving the LM curve where it was. Expected deflation therefore explains a fall in both nominal and real income as well as in the nominal rate of interest; it also explains an increase in the real rate of interest. Scott seems to be emotionally committed to the notion that the IS-LM model must lead to a misunderstanding of the effects of monetary policy, holding Foote up as an example of this confusion on the basis of the first of the slides, but Foote actually shows that IS-LM can be tweaked to accommodate a correct understanding of the dominant role of monetary policy in the Great Depression.

The Great Depression was triggered by a deflationary scramble for gold associated with the uncoordinated restoration of the gold standard by the major European countries in the late 1920s, especially France and its insane central bank. On top of this, the Federal Reserve, succumbing to political pressure to stop “excessive” stock-market speculation, raised its discount rate to a near record 6.5% in early 1929, greatly amplifying the pressure on gold reserves, thereby driving up the value of gold, and causing expectations of the future price level to start dropping. It was thus a rise (both actual and expected) in the value of gold, not a reduction in the money supply, which was the source of the monetary shock that produced the Great Depression. The shock was administered without a reduction in the money supply, so there was no shift in the LM curve. IS-LM is not necessarily the best model with which to describe this monetary shock, but the basic story can be expressed in terms of the IS-LM model.

So, you ask, if I don’t think that Foote’s exposition of the IS-LM model seriously misrepresents what happened in the Great Depression, why did I say at beginning of this post that Foote’s slides really annoy me? Well, the reason is simply that Foote seems to think that the only monetary explanation of the Great Depression is the Monetarist explanation of Milton Friedman: that the Great Depression was caused by an exogenous contraction in the US money supply. That explanation is wrong, theoretically and empirically.

What caused the Great Depression was an international disturbance to the value of gold, caused by the independent actions of a number of central banks, most notably the insane Bank of France, maniacally trying to convert all its foreign exchange reserves into gold, and the Federal Reserve, obsessed with suppressing a non-existent stock-market bubble on Wall Street. It only seems like a bubble with mistaken hindsight, because the collapse of prices was not the result of any inherent overvaluation in stock prices in October 1929, but because the combined policies of the insane Bank of France and the Fed wrecked the world economy. The decline in the nominal quantity of money in the US, the great bugaboo of Milton Friedman, was merely an epiphenomenon.

As Ron Batchelder and I have shown, Gustav Cassel and Ralph Hawtrey had diagnosed and explained the causes of the Great Depression fully a decade before it happened. Unfortunately, whenever people think of a monetary explanation of the Great Depression, they think of Milton Friedman, not Hawtrey and Cassel. Scott Sumner understands all this, he’s even written a book – a wonderful (but unfortunately still unpublished) book – about it. But he gets all worked up about IS-LM.

I, on the other hand, could not care less about IS-LM; it’s the idea that the monetary cause of the Great Depression was discovered by Milton Friedman that annoys the [redacted] out of me.

UPDATE: I posted this post prematurely before I finished editing it, so I apologize for any mistakes or omissions or confusing statements that appeared previously or that I haven’t found yet.

The Backing Theory of Money v. the Quantity Theory of Money

Mike Sproul and Scott Sumner were arguing last week about how to account for the value of fiat money and the rate of inflation. As I observed in a recent post, I am doubtful that monetary theory, in its current state, can handle those issues adequately, so I am glad to see that others are trying to think the problems through even if the result is only to make clear how much we don’t know. Both Mike and Scott are very smart guys, and I find some validity in the arguments of both even if I am not really satisfied with the arguments of either.

Mike got things rolling with a guest post on JP Koning’s blog in which he lodged two complaints against Scott:

First, “Scott thinks that the liabilities of governments and central banks are not really liabilities.”

I see two problems with Mike’s first complaint. First, Mike is not explicit about which liabilities he is referring to. However, from the context of his discussion, it seems clear that he is talking about those liabilities that we normally call currency, or in the case of the Federal Reserve, Federal Reserve Notes. Second, and more important, it is not clear what definition of “liability” Mike is using. In a technical sense, as Mike observes, Federal Reserve Notes are classified by the Fed itself as liabilities. But what does it mean for a Federal Reserve Note to be a liability of the Fed? A liability implies that an obligation has been undertaken by someone to be discharged under certain defined conditions. What is the obligation undertaken by the Fed upon issuing a Federal Reserve Note. Under the gold standard, the Fed was legally obligated to redeem its Notes for gold at a fixed predetermined conversion rate. After the gold standard was suspended, that obligation was nullified. What obligation did the Fed accept in place of the redemption obligation? Here’s Mike’s answer:

But there are at least three other ways that FRN’s can still be redeemed: (i) for the Fed’s bonds, (ii) for loans made by the Fed, (iii) for taxes owed to the federal government. The Fed closed one channel of redemption (the gold channel), while the other redemption channels (loan, tax, and bond) were left open.

Those are funny obligations inasmuch as there are no circumstances under which they require the Fed to take any action. The purchase of a Fed (Treasury?) bond at the going market price imposes no obligation on the Fed to do anything except what it is already doing anyway. For there to be an obligation resulting from the issue by the Fed of a note, it would have been necessary for the terms of the transaction following upon the original issue to have been stipulated in advance. But the terms on which the Fed engages in transactions with the public are determined by market forces not by contractual obligation. The same point applies to loans made by the Fed. When the Fed makes a loan, it emits FRNs. The willingness of the Fed to accept FRNs previously emitted in the course of making loans as repayment of those loans doesn’t strike me as an obligation associated with its issue of FRNs. Finally, the fact that the federal government accepts (or requires) payment of tax obligations in FRNs is a decision of the Federal government to which the Fed as a matter of strict legality is not a party. So it seems to me that the technical status of an FRN as a liability of the Fed is a semantic or accounting oddity rather than a substantive property of a FRN.

Having said that, I think that Mike actually does make a substantive point about FRNs, which is that FRNs are not necessarily hot potatoes in the strict quantity-theory sense. There are available channels through which the public can remit its unwanted FRNs back to the Fed. The economic question is whether those means of sending unwanted FRNs back to the Fed are as effective in pinning down the price level as an enforceable legal obligation undertaken by the Fed to redeem FRNs at a predetermined exchange rate in terms of gold. Mike suggests that the alternative mechanisms by which the public can dispose of unwanted FRNs are as effective as gold convertibility in pinning down the price level. I think that assertion is implausible, and it remains to be proved, though I am willing to keep an open mind on the subject.

Now let’s consider Mike’s second complaint: “Scott thinks that if the central bank issues more money, then the money will lose value even if the money is fully backed.”

My first reaction is to ask what it means for money to be “fully backed?” Since it is not clear in what sense the inconvertible note issue of a central bank represents a liability of the issuing bank, it is also not exactly clear why any backing is necessary, or what backing means, though I will try to suggest in a moment a reason why the assets of the central bank actually do matter. But again the point is that, when a liability does not impose a well-defined legal obligation on the central bank to redeem that liability at a predetermined rate in terms of an asset whose supply the central bank does not itself control, the notion of “backing” is as vague as the notion of a “liability.” The difference between a liability that imposes no effective constraint on a central bank and one that does impose an effective constraint on a central bank is the difference between what Nick Rowe calls an alpha bank, which does not make its notes convertible into another asset (real or monetary) not under its control, and what he calls a beta bank, which does make its liabilities convertible into another asset (real or monetary) not under its control.

Now one way to interpret “backing” is to look at all the assets on the balance sheet of the central bank and compare the value of those assets to the value of the outstanding notes issued by the central bank. Sometimes I think that this is really all that Mike means when he talks about “backing,” but I am not really sure. At any rate, if we think of backing in this vague sense, maybe what Mike wants to say is that the value of the outstanding note issue of the central bank is equal to the value of its assets divided by the amount of notes that it has issued. But if this really is what Mike means, then it seems that the aggregate value of the outstanding notes of the central bank must always equal the value of the assets of the central bank. But there is a problem with that notion of “backing” as well, because the equality in the value of the assets of the central bank and its liabilities can be achieved at any price level, and at any rate of inflation, because an increase in prices will scale up the nominal value of outstanding notes and the value of central-bank assets by the same amount. Without providing some nominal anchor, which, as far as I can tell, Mike has not done, the price level is indeterminate. Now to be sure, this is no reason for quantity theorist like Scott to feel overly self-satisfied, because the quantity theory is subject to the same indeterminacy. And while Mike seems absolutely convinced that the backing theory is superior to the quantity theory, he himself admits that it is very difficult, if not impossible, to distinguish between the two theories in terms of their empirical implications.

Let me now consider a slightly different way in which the value of the assets on the balance sheet of a central bank could affect the value of the money issued by the central bank. I would suggest, along the lines of an argument made by Ben Klein many years ago in some of his papers on competitive moneys (e.g. this one), that it is meaningful to talk about the quality of the money issued by a particular bank. In Klein’s terms, the quality of a money reflects the confidence with which people can predict the future value of a money. It’s plausible to assume that the demand (in real terms) to hold money increases with the quality of money. Certainly people will tend to switch form holding lower- to higher-quality moneys. I think that it’s also plausible to assume that the quality of a particular money issued by a central bank increases as the value of the assets held by the central bank increases, because the larger the asset portfolio of the issuer, the more likely it is that the issuer will control the value of the money that it has issued. (This goes to Mike’s point that a central bank has to hold enough assets to buy back its currency if the demand for it goes down. Actually it doesn’t, but people will be more willing to hold a money the larger the stock of assets held by the issuer with which it can buy back its money to prevent it from losing value.) I think that is ultimately the idea that Mike is trying to get at when he talks about “backing.” So I would interpret Mike as saying that the quality of a money is an increasing function of the total asset holdings of the central bank issuing the money, and the demand for a money is an increasing function of its quality. Such an adjustment in Mike’s backing theory just might help to bring the backing theory and the quantity theory into a closer correspondence than one might gather from reading the back and forth between Mike and Scott last week.

PS Mike was kind enough to quote my argument about the problem that backward induction poses for the standard explanation of the value of fiat money. Scott once again dismisses the problem by saying that the problem can be avoided by assuming that no one knows when the last period is. I agree that that is a possible answer, but it means that the value of fiat money is contingent on a violation of rational expectations and the efficient market hypothesis. I am sort of surprised that Scott, of all people, would be so nonchalant about accepting such a violation. But I’ve already said enough about that for now.

Never Reason from a Disequilibrium

One of Scott Sumner’s many contributions as a blogger has been to show over and over and over again how easy it is to lapse into fallacious economic reasoning by positing a price change and then trying to draw inferences about the results of the price change. The problem is that a price change doesn’t just happen; it is the result of some other change. There being two basic categories of changes (demand and supply) that can affect price, there are always at least two possible causes for a given price change. So, until you have specified the antecedent change responsible for the price change under consideration, you can’t work out the consequences of the price change.

In this post, I want to extend Scott’s insight in a slightly different direction, and explain how every economic analysis has to begin with a statement about the initial conditions from which the analysis starts. In particular, you need to be clear about the equilibrium position corresponding to the initial conditions from which you are starting. If you posit some change in the system, but your starting point isn’t an equilibrium, you have no way of separating out the adjustment to the change that you are imposing on the system from the change the system would be undergoing simply to reach the equilibrium toward which it is already moving, or, even worse, from the change the system would be undergoing if its movement is not toward equilibrium.

Every theoretical analysis in economics properly imposes a ceteris paribus condition. Unfortunately, the ubiquitous ceteris paribus condition comes dangerously close to rendering economic theory irrefutable, except perhaps in a statistical sense, because empirical refutations of the theory can always be attributed to changes, abstracted from only in the theory, but not in the real world of our experience. An empirical model with a sufficient number of data points may be able to control for the changes in conditions that the theory holds constant, but the underlying theory is a comparison of equilibrium states (comparative statics), and it is quite a stretch to assume that the effects of perpetual disequilibrium can be treated as nothing but white noise. Austrians are right to be skeptical of econometric analysis; so was Keynes, for that matter. But skepticism need not imply nihilism.

Let me try to illustrate this principle by applying it to the Keynesian analysis of involuntary unemployment. In the General Theory Keynes argued that if adequate demand is deficient, the likely result is an equilibrium with involuntary unemployment. The “classical” argument that Keynes disputed was that, in principle at least, involuntary unemployment could not persist, because unemployed workers, if only they would accept reduced money wages, would eventually find employment. Keynes denied that involuntary unemployment could not persist, arguing that if workers did accept reduced money wages, the wage reductions would not get translated into reduced real wages. Instead, falling nominal wages would induce employers to cut prices by roughly the same percentage as the reduction in nominal wages, leaving real wages more or less unchanged, thereby nullifying the effectiveness of nominal-wage cuts, and, instead, fueling a vicious downward spiral of prices and wages.

In making this argument, Keynes didn’t dispute the neoclassical proposition that, with a given capital stock, the marginal product of labor declines as employment increases, implying that real wages have to fall for employment to be increased. His argument was about the nature of the labor-supply curve, labor supply, in Keynes’s view, being a function of both the real and the nominal wage, not, as in the neoclassical theory, only the real wage. Under Keynes’s “neoclassical” analysis, the problem with nominal-wage cuts is that they don’t do the job, because they lead to corresponding price cuts. The only way to reduce unemployment, Keynes insisted, is to raise the price level. With nominal wages constant, an increased price level would achieve the real-wage cut necessary for employment to be increased. And this is precisely how Keynes defined involuntary unemployment: the willingness of workers to increase the amount of labor actually supplied in response to a price level increase that reduces their real wage.

Interestingly, in trying to explain why nominal-wage cuts would fail to increase employment, Keynes suggested that the redistribution of income from workers to entrepreneurs associated with reduced nominal wages would tend to reduce consumption, thereby reducing, not increasing, employment. But if that is so, how is it that a reduced real wage, achieved via inflation, would increase employment? Why would the distributional effect of a reduced nominal, but unchanged real, wage be more adverse to employment han a reduced real wage, achieved, with a fixed nominal wage, by way of a price-level increase?

Keynes’s explanation for all this is confused. In chapter 19, where he makes the argument that money-wage cuts can’t eliminate involuntary unemployment, he presents a variety of reasons why nominal-wage cuts are ineffective, and it is usually not clear at what level of theoretical abstraction he is operating, and whether he is arguing that nominal-wage cuts would not work even in principle, or that, although nominal-wage cuts might succeed in theory, they would inevitably fail in practice. Even more puzzling, It is not clear whether he thinks that real wages have to fall to achieve full employment or that full employment could be restored by an increase in aggregate demand with no reduction in real wages. In particular, because Keynes doesn’t start his analysis from a full-employment equilibrium, and doesn’t specify the shock that moves the economy off its equilibrium position, we can only guess whether Keynes is talking about a shock that had reduced labor productivity or (more likely) a shock to entrepreneurial expectations (animal spirits) that has no direct effect on labor productivity.

There was a rhetorical payoff for Keynes in maintaining that ambiguity, because he wanted to present a “general theory” in which full employment is a special case. Keynes therefore emphasized that the labor market is not self-equilibrating by way of nominal-wage adjustments. That was a perfectly fine and useful insight: when the entire system is out of kilter; there is no guarantee that just letting the free market set prices will bring everything back into place. The theory of price adjustment is fundamentally a partial-equilibrium theory that isolates the disequiibrium of a single market, with all other markets in (approximate) equilibrium. There is no necessary connection between the adjustment process in a partial-equilibrium setting and the adjustment process in a full-equilibrium setting. The stability of a single market in disequilibrium does not imply the stability of the entire system of markets in disequilibrium. Keynes might have presented his “general theory” as a theory of disequilibrium, but he preferred (perhaps because he had no other tools to work with) to spell out his theory in terms of familiar equilibrium concepts: savings equaling investment and income equaling expenditure, leaving it ambiguous whether the failure to reach a full-employment equilibrium is caused by a real wage that is too high or an interest rate that is too high. Axel Leijonhufvud highlights the distinction between a disequilibrium in the real wage and a disequilibrium in the interest rate in an important essay “The Wicksell Connection” included in his book Information and Coordination.

Because Keynes did not commit himself on whether a reduction in the real wage is necessary for equilibrium to be restored, it is hard to assess his argument about whether, by accepting reduced money wages, workers could in fact reduce their real wages sufficiently to bring about full employment. Keynes’s argument that money-wage cuts accepted by workers would be undone by corresponding price cuts reflecting reduced production costs is hardly compelling. If the current level of money wages is too high for firms to produce profitably, it is not obvious why the reduced money wages paid by entrepreneurs would be entirely dissipated by price reductions, with none of the cost decline being reflected in increased profit margins. If wage cuts do increase profit margins, that would encourage entrepreneurs to increase output, potentially triggering an expansionary multiplier process. In other words, if the source of disequilibrium is that the real wage is too high, the real wage depending on both the nominal wage and price level, what is the basis for concluding that a reduction in the nominal wage would cause a change in the price level sufficient to keep the real wage at a disequilibrium level? Is it not more likely that the price level would fall no more than required to bring the real wage back to the equilibrium level consistent with full employment? The question is not meant as an expression of policy preference; it is a question about the logic of Keynes’s analysis.

Interestingly, present-day opponents of monetary stimulus (for whom “Keynesian” is a term of extreme derision) like to make a sort of Keynesian argument. Monetary stimulus, by raising the price level, reduces the real wage. That means that monetary stimulus is bad, as it is harmful to workers, whose interests, we all know, is the highest priority – except perhaps the interests of rentiers living off the interest generated by their bond portfolios — of many opponents of monetary stimulus. Once again, the logic is less than compelling. Keynes believed that an increase in the price level could reduce the real wage, a reduction that, at least potentially, might be necessary for the restoration of full employment.

But here is my question: why would an increase in the price level reduce the real wage rather than raise money wages along with the price level. To answer that question, you need to have some idea of whether the current level of real wages is above or below the equilibrium level. If unemployment is high, there is at least some reason to think that the equilibrium real wage is less than the current level, which is why an increase in the price level would be expected to cause the real wage to fall, i.e., to move the actual real wage in the direction of equilibrium. But if the current real wage is about equal to, or even below, the equilibrium level, then why would one think that an increase in the price level would not also cause money wages to rise correspondingly? It seems more plausible that, in the absence of a good reason to think otherwise, that inflation would cause real wages to fall only if real wages are above their equilibrium level.

Did Raising Interest Rates under the Gold Standard Really Increase Aggregate Demand?

I hope that I can write this quickly just so people won’t think that I’ve disappeared. I’ve been a bit under the weather this week, and the post that I’ve been working on needs more attention and it’s not going to be ready for a few more days. But the good news, from my perspective at any rate, is that Scott Sumner, as he has done so often in the past, has come through for me by giving me something to write about. In his most recent post at his second home on Econlog, Scott writes the following:

I recently did a post pointing out that higher interest rates don’t reduce AD.  Indeed even higher interest rates caused by a decrease in the money supply don’t reduce AD. Rather the higher rates raise velocity, but that effect is more than offset by the decrease in the money supply.

Of course that’s not the way Keynesians typically look at things.  They believe that higher interest rates actually cause AD to decrease.  Except under the gold standard. Back in 1988 Robert Barsky and Larry Summers wrote a paper showing that higher interest rates were expansionary when the dollar was pegged to gold.  Now in fairness, many Keynesians understand that higher interest rates are often associated with higher levels of AD.  But Barsky and Summers showed that the higher rates actually caused AD to increase.  Higher nominal rates increase the opportunity cost of holding gold. This reduces gold demand, and thus lowers its value.  Because the nominal price of gold is fixed under the gold standard, the only way for the value of gold to decrease is for the price level to increase. Thus higher interest rates boost AD and the price level.  This explains the “Gibson Paradox.”

Very clever on Scott’s part, and I am sure that he will have backfooted a lot of Keynesians. There’s just one problem with Scott’s point, which is that he forgets that an increase in interest rates by the central bank under the gold standard corresponds to an increase in the demand of the central bank for gold, which, as Scott certainly knows better than almost anyone else, is deflationary. What Barsky and Summers were talking about when they were relating interest rates to the value of gold was movements in the long-term interest rate (the yield on consols), not in central-bank lending rate (the rate central banks charge for overnight or very short-dated loans to other banks). As Hawtrey showed in A Century of Bank Rate, the yield on consols was not closely correlated with Bank Rate. So not only is Scott looking at the wrong interest rate (for purposes of his argument), he is – and I don’t know how to phrase this delicately – reasoning from a price change. Ouch!

Never Mistake a Change in Quantity Demanded for a Change in Demand

We are all in Scott Sumner’s debt for teaching (or reminding) us never, ever to reason from a price change. The reason is simple. You can’t just posit a price change and then start making inferences from the price change, because price changes don’t just happen spontaneously. If there’s a price change, it’s because something else has caused price to change. Maybe demand has increased; maybe supply has decreased; maybe neither supply nor demand has changed, but the market was in disequilibrium before and is in equilibrium now at the new price; maybe neither supply nor demand has changed, but the market was in equilibrium before and is in disequilibrium now. There could be other scenarios as well, but unless you specify at least one of them, you can’t reason sensibly about the implications of the price change.

There’s another important piece of advice for anyone trying to do economics: never mistake a change in quantity demanded for a change in demand. A change in demand means that the willingness of people to pay for something has changed, so that, everything else held constant, the price has to change. If for some reason, the price of something goes up, the willingness of people to pay for not having changed, then the quantity of the thing that they demand will go down. But here’s the important point: their demand for that something – their willingness to pay for it – has not gone down; the change in the amount demanded is simply a response to the increased price of that something. In other words, a change in the price of something cannot be the cause of a change in the demand for that something; it can only cause a change in the quantity demanded. A change in demand can be caused only by change in something other than price – maybe a change in wealth, or in fashion, or in taste, or in the season, or in the weather.

Why am I engaging in this bit of pedantry? Well, in a recent post, Scott responded to the following question from Dustin in the comment section to one of his posts.

An elementary question on the topic of interest rates that I’ve been unable to resolve via google:

Regarding Fed actions, I understand that reduced interest rates are thought to be expansionary because the resulting decrease in cost of capital induces greater investment. But I also understand that reduced interest rates are thought to be contractionary because the resulting decrease in opportunity cost of holding money increases demand for money.

To which Scott responded as follows:

It’s not at all clear that lower interest rates boost investment (never reason from a price change.)  And even if they did boost investment it is not at all clear that they would boost GDP.

Scott is correct to question the relationship between interest rates and investment. The relationship in the Keynesian model is based on the idea that a reduced interest rate, by reducing the rate at which expected future cash flows are discounted, increases the value of durable assets, so that the optimal size of the capital stock increases, implying a speed up in the rate of capital accumulation (investment). There are a couple of steps missing in the chain of reasoning that goes from a reduced rate of discount to a speed up in the rate of accumulation, but, in the olden days at any rate, economists have usually been willing to rely on their intuition that an increase in the size of the optimal capital stock would translate into an increased rate of capital accumulation.

Alternatively, in the Hawtreyan scheme of things, a reduced rate of interest would increase the optimal size of inventories held by traders and middlemen, causing an increase in orders to manufacturers, and a cycle of rising output and income generated by the attempt to increase inventories. Notice that in the Hawtreyan view, the reduced short-term interest is, in part, a positive supply shock (reducing the costs borne by middlemen and traders of holding inventories financed by short-term borrowing) as long as there are unused resources that can be employed if desired inventories increase in size.

That said, I’m not sure what Scott, in questioning whether a reduction in interesting rates raises investment, meant by his parenthetical remark about reasoning from a price change. Scott was asked about the effect of a Fed policy to reduce interest rates. Why is that reasoning from a price change? And furthermore, if we do posit that investment rises, why is it unclear whether GDP would rise?

Scott continues:

However it’s surprisingly hard to explain why OMPs boost NGDP using the mechanism of interest rates. Dustin is right that lower interest rates increase the demand for money.  They also reduce velocity. Higher money demand and lower velocity will, ceteris paribus, reduce NGDP.  So why does everyone think that a cut in interest rates increases NGDP?  Is it possible that Steve Williamson is right after all?

Sorry, Scott. Lower interest rates don’t increase the demand for money; lower interest rates increase the amount of money demanded. What’s the difference? If an interest-rate reduction increased the demand for money, it would mean that the demand curve had shifted, and the size of that shift would be theoretically unspecified. If that were the case, we would be comparing an unknown increase in investment on the one hand to an unknown increase in money demand on the other hand, the net effect being indeterminate. That’s the argument that Scott seems to be making.

But that’s not, repeat not, what’s going on here. What we have is an interest-rate reduction that triggers an increase investment and also in the amount of money demanded. But there is no shift in the demand curve for money, just a movement along an unchanging demand curve. That imposes a limit on the range of possibilities. What is the limit? It’s the extreme case of a demand curve for money that is perfectly elastic at the current rate of interest — in other words a liquidity trap — so that the slightest reduction in interest rates causes an unlimited increase in the amount of money demanded. But that means that the rate of interest can’t fall, so that investment can’t rise. If the demand for money is less than perfectly elastic, then the rate of interest can in fact be reduced, implying that investment, and therefore NGDP, will increase. The quantity of money demanded increases as well — velocity goes down — but not enough to prevent investment and NGDP from increasing.

So, there’s no ambiguity about the correct answer to Dustin’s question. If Steve Williamson is right, it’s because he has introduced some new analytical element not contained in the old-fashioned macroeconomic analysis. (Note that I use the term “old-fashioned” only as an identifier, not as an expression of preference in either direction.) A policy-induced reduction in the rate of interest must, under standard assumption in the old-fashioned macroeconomics, increase nominal GDP, though the size of the increase depends on specific assumptions about empirical magnitudes. I don’t disagree with Scott’s analysis in terms of the monetary base, I just don’t see a substantive difference between that analysis and the one that I just went through in terms of the interest-rate policy instrument.

Just to offer a non-controversial example, it is possible to reason through the effect of a restriction on imports in terms of a per unit tariff on imports or in terms of a numerical quota on imports. For any per unit tariff, there is a corresponding quota on imports that gives you the same solution. MMT guys often fail to see the symmetry between setting the quantity and the price of bank reserves; in this instance Scott seems to have overlooked the symmetry between the quantity and price of base money.

Does Macroeconomics Need Financial Foundations?

One of the little instances of collateral damage occasioned by the hue and cry following upon Stephen Williamson’s post arguing that quantitative easing has been deflationary was the dustup between Scott Sumner and financial journalist and blogger Izabella Kaminska. I am not going to comment on the specifics of their exchange except to say that the misunderstanding and hard feelings between them seem to have been resolved more or less amicably. However, in quickly skimming the exchange between them, I was rather struck by the condescending tone of Kaminska’s (perhaps understandable coming from the aggrieved party) comment about the lack of comprehension by Scott and Market Monetarists more generally of the basics of finance.

First I’d just like to say I feel much of the misunderstanding comes from the fact that market monetarists tend to ignore the influence of shadow banking and market plumbing in the monetary world. I also think (especially from my conversation with Lars Christensen) that they ignore technological disruption, and the influence this has on wealth distribution and purchasing decisions amongst the wealthy, banks and corporates. Also, as I outlined in the post, my view is slightly different to Williamson’s, it’s based mostly on the scarcity of safe assets and how this can magnify hoarding instincts and fragment store-of-value markets, in a Gresham’s law kind of way. Expectations obviously factor into it, and I think Williamson is absolutely right on that front. But personally I don’t think it’s anything to do with temporary or permanent money expansion expectations. IMO It’s much more about risk expectations, which can — if momentum builds — shift very very quickly, making something deflationary, inflationary very quickly. Though, that doesn’t mean I am worried about inflation (largely because I suspect we may have reached an important productivity inflection point).

This remark was followed up with several comments blasting Market Monetarists for their ignorance of the basics of finance and commending Kaminska for the depth of her understanding to which Kaminska warmly responded adding a few additional jibes at Sumner and Market Monetarists. Here is one.

Market monetarists are getting testy because now that everybody started scrutinizing QE they will be exposed as ignorant. The mechanisms they originally advocated QE would work through will be seen as hopelessly naive. For them the money is like glass beads squirting out of the Federal Reserve, you start talking about stuff like collateral, liquid assets, balance sheets and shadow banking and they are out of their depth.

For laughs: Sumner once tried to defend the childish textbook model of banks lending out reserves and it ended in a colossal embarrassment in the comments section http://www.themoneyillusion.com/?p=5893

For you to defend your credentials in front of such “experts” is absurd. There is a lot more depth to your understanding than to their sandbox vision of the monetary system. And yes, it *is* crazy that journalists and bloggers can talk about these things with more sense than academics. But this [is] the world we live in.

To which Kaminska graciously replied:

Thanks as well! And I tend to agree with your assessment of the market monetarist view of the world.

So what is the Market Monetarist view of the world of which Kaminska tends to have such a low opinion? Well, from reading Kaminska’s comments and those of her commenters, it seems to be that Market Monetarists have an insufficiently detailed and inaccurate view of financial intermediaries, especially of banks and shadow banks, and that Market Monetarists don’t properly understand the role of safe assets and collateral in the economy. But the question is why, and how, does any of this matter to a useful description of how the economy works?

Well, this whole episode started when Stephen Williamson had a blog post arguing that QE was deflationary, and the reason it’s deflationary is that creating more high powered money provides the economy with more safe assets and thereby reduces the liquidity premium associated with safe assets like short-term Treasuries and cash. By reducing the liquidity premium, QE causes the real interest rate to fall, which implies a lower rate of inflation.

Kaminska thinks that this argument, which Market Monetarists find hard to digest, makes sense, though she can’t quite bring herself to endorse it either. But she finds the emphasis on collateral and safety and market plumbing very much to her taste. In my previous post, I raised what I thought were some problems with Williamson’s argument.

First, what is the actual evidence that there is a substantial liquidity premium on short-term Treasuries? If I compare the rates on short-term Treasuries with the rates on commercial paper issued by non-Financial institutions, I don’t find much difference. If there is a substantial unmet demand for good collateral, and there is only a small difference in yield between commercial paper and short-term Treasuries, one would think that non-financial firms could make a killing by issuing a lot more commercial paper. When I wrote the post, I was wondering whether I, a financial novice, might be misreading the data or mismeasuring the liquidity premium on short-term Treasuries. So far, no one has said anything about that, but If I am wrong, I am happy to be enlightened.

Here’s something else I don’t get. What’s so special about so-called safe assets? Suppose, as Williamson claims, that there’s a shortage of safe assets. Why does that imply a liquidity premium? One could still compensate for the lack of safety by over-collateralizing the loan using an inferior asset. If that is a possibility, why is the size of the liquidity premium not constrained?

I also pointed out in my previous post that a declining liquidity premium would be associated with a shift out of money and into real assets, which would cause an increase in asset prices. An increase in asset prices would tend to be associated with an increase in the value of the underlying service flows embodied in the assets, in other words in an increase in current prices, so that, if Williamson is right, QE should have caused measured inflation to rise even as it caused inflation expectations to fall. Of course Williamson believes that the decrease in liquidity premium is associated with a decline in real interest rates, but it is not clear that a decline in real interest rates has any implications for the current price level. So Williamson’s claim that his model explains the decline in observed inflation since QE was instituted does not seem all that compelling.

Now, as one who has written a bit about banking and shadow banking, and as one who shares the low opinion of the above-mentioned commenter on Kaminska’s blog about the textbook model (which Sumner does not defend, by the way) of the money supply via a “money multiplier,” I am in favor of changing how the money supply is incorporated into macromodels. Nevertheless, it is far from clear that changing the way that the money supply is modeled would significantly change any important policy implications of Market Monetarism. Perhaps it would, but if so, that is a proposition to be proved (or at least argued), not a self-evident truth to be asserted.

I don’t say that finance and banking are not important. Current spreads between borrowing and lending rates, may not provide a sufficient margin for banks to provide the intermediation services that they once provided to a wide range of customers. Businesses have a wider range of options in obtaining financing than they used to, so instead of holding bank accounts with banks and foregoing interest on deposits to be able to have a credit line with their banker, they park their money with a money market fund and obtain financing by issuing commercial paper. This works well for firms large enough to have direct access to lenders, but smaller businesses can’t borrow directly from the market and can only borrow from banks at much higher rates or by absorbing higher costs on their bank accounts than they would bear on a money market fund.

At any rate, when market interest rates are low, and when perceived credit risks are high, there is very little margin for banks to earn a profit from intermediation. If so, the money multiplier — a crude measure of how much intermediation banks are engaging in goes down — it is up to the monetary authority to provide the public with the liquidity they demand by increasing the amount of bank reserves available to the banking system. Otherwise, total spending would contract sharply as the public tried to build up their cash balances by reducing their own spending – not a pretty picture.

So finance is certainly important, and I really ought to know more about market plumbing and counterparty risk  and all that than I do, but the most important thing to know about finance is that the financial system tends to break down when the jointly held expectations of borrowers and lenders that the loans that they agreed to would be repaid on schedule by the borrowers are disappointed. There are all kinds of reasons why, in a given case, those jointly held expectations might be disappointed. But financial crises are associated with a very large cluster of disappointed expectations, and try as they might, the finance guys have not provided a better explanation for that clustering of disappointed expectations than a sharp decline in aggregate demand. That’s what happened in the Great Depression, as Ralph Hawtrey and Gustav Cassel and Irving Fisher and Maynard Keynes understood, and that’s what happened in the Little Depression, as Market Monetarists, especially Scott Sumner, understand. Everything else is just commentary.

The Internal Contradiction of Quantitative Easing

Last week I was struggling to cut and paste my 11-part series on Hawtrey’s Good and Bad Trade into the paper on that topic that I am scheduled to present next week at the Southern Economic Association meetings in Tampa Florida, completing the task just before coming down with a cold which has kept me from doing anything useful since last Thursday. But I was at least sufficiently aware of my surroundings to notice another flurry of interest in quantitative easing, presumably coinciding with Janet Yellen’s testimony at the hearings conducted by the Senate Banking Committee about her nomination to succeed Ben Bernanke as Chairman of Federal Reserve Board.

In my cursory reading of the latest discussions, I didn’t find see a lot that has not already been said, so I will take that as an opportunity to restate some points that I have previously made on this blog. But before I do that, I can’t help observing (not for the first time either) that the two main arguments made by critics of QE do not exactly coexist harmoniously with each other. First, QE is ineffective; second it is dangerous. To be sure, the tension between these two claims about QE does not prove that both can’t be true, and certainly doesn’t prove that both are wrong. But the tension might at least have given a moment’s pause to those crying that Quantitative Easing, having failed for five years to accomplish anything besides enriching Wall Street and taking bread from the mouths of struggling retirees, is going to cause the sky to fall any minute.

Nor, come to think of it, does the faux populism of the attack on a rising stock market and of the crocodile tears for helpless retirees living off the interest on their CDs coexist harmoniously with the support by many of the same characters opposing QE (e.g., Freedomworks, CATO, the Heritage Foundation, and the Wall Street Journal editorial page) for privatizing social security via private investment accounts to be invested in the stock market, the argument being that the rate of return on investing in stocks has historically been greater than the rate of return on payments into the social security system. I am also waiting for an explanation of why abused pensioners unhappy with the returns on their CDs can’t cash in the CDs and buy dividend-paying-stocks? In which charter of the inalienable rights of Americans, I wonder, does one find it written that a perfectly secure real rate of interest of not less than 2% on any debt instrument issued by the US government shall always be guaranteed?

Now there is no denying that what is characterized as a massive program of asset purchases by the Federal Reserve System has failed to stimulate a recovery comparable in strength to almost every recovery since World War II. However, not even the opponents of QE are suggesting that the recovery has been weak as a direct result of QE — that would be a bridge too far even for the hard money caucus — only that whatever benefits may have been generated by QE are too paltry to justify its supposedly bad side-effects (present or future inflation, reduced real wages, asset bubbles, harm to savers, enabling of deficit-spending, among others). But to draw any conclusion about the effects of QE, you need some kind of a baseline of comparison. QE opponents therefore like to use previous US recoveries, without the benefit of QE, as their baseline.

But that is not the only baseline available for purposes of comparison. There is also the Eurozone, which has avoided QE and until recently kept interest rates higher than in the US, though to be sure not as high as US opponents of QE (and defenders of the natural rights of savers) would have liked. Compared to the Eurozone, where nominal GDP has barely risen since 2010, and real GDP and employment have shrunk, QE, which has been associated with nearly 4% annual growth in US nominal GDP and slightly more than 2% annual growth in US real GDP, has clearly outperformed the eurozone.

Now maybe you don’t like the Eurozone, as it includes all those dysfunctional debt-ridden southern European countries, as a baseline for comparison. OK, then let’s just do a straight, head-to-head matchup between the inflation-addicted US and solid, budget-balancing, inflation-hating Germany. Well that comparison shows (see the chart below) that since 2011 US real GDP has increased by about 5% while German real GDP has increased by less than 2%.

US_Germany_RGDP

So it does seem possible that, after all, QE and low interest rates may well have made things measurably better than they would have otherwise been. But don’t expect to opponents of QE to acknowledge that possibility.

Of course that still leaves the question on the table, why has this recovery been so weak? Well, Paul Krugman, channeling Larry Summers, offered a demographic hypothesis in his column Monday: that with declining population growth, there have been diminishing investment opportunities, which, together with an aging population, trying to save enough to support themselves in their old age, causes the supply of savings to outstrip the available investment opportunities, driving the real interest rate down to zero. As real interest rates fall, the ability of the economy to tolerate deflation — or even very low inflation — declines. That is a straightforward, and inescapable, implication of the Fisher equation (see my paper “The Fisher Effect Under Deflationary Expectations”).

So, if Summers and Krugman are right – and the trend of real interest rates for the past three decades is not inconsistent with their position – then we need to rethink revise upwards our estimates of what rate of inflation is too low. I will note parenthetically, that Samuel Brittan, who has been for decades just about the most sensible economic journalist in the world, needs to figure out that too little inflation may indeed be a bad thing.

But this brings me back to the puzzling question that causes so many people to assume that monetary policy is useless. Why have trillions of dollars of asset purchases not generated the inflation that other monetary expansions have generated? And if all those assets now on the Fed balance sheet haven’t generated inflation, what reason is there to think that the Fed could increase the rate of inflation if that is what is necessary to avoid chronic (secular) stagnation?

The answer, it seems to me is the following. If everyone believes that the Fed is committed to its inflation target — and not even the supposedly dovish Janet Yellen, bless her heart, has given the slightest indication that she favors raising the Fed’s inflation target, a target that, recent experience shows, the Fed is far more willing to undershoot than to overshoot – then Fed purchases of assets with currency are not going to stimulate additional private spending. Private spending, at or near the zero lower bound, are determined largely by expectations of future income and prices. The quantity of money in private hands, being almost costless to hold, is no longer a hot potato. So if there is no desire to reduce excess cash holdings, the only mechanism by which monetary policy can affect private spending is through expectations. But the Fed, having succeeded in anchoring inflation expectations at 2%, has succeeded in unilaterally disarming itself. So economic expansion is constrained by the combination of a zero real interest rate and expected inflation held at or below 2% by a political consensus that the Fed, even if it were inclined to, is effectively powerless to challenge.

Scott Sumner calls this monetary offset. I don’t think that we disagree much on the economic analysis, but it seems to me that he overestimates the amount of discretion that the Fed can actually exercise over monetary policy. Except at the margins, the Fed is completely boxed in by a political consensus it dares not question. FDR came into office in 1933, and was able to effect a revolution in monetary policy within his first month in office, thereby saving the country and Western Civilization. Perhaps Obama had an opportunity to do something similar early in his first term, but not any more. We are stuck at 2%, but it is no solution.

Friedman’s Dictum

In his gallant, but in my opinion futile, attempts to defend Milton Friedman against the scandalous charge that Friedman was, gasp, a Keynesian, if not in his policy prescriptions, at least in his theoretical orientation, Scott Sumner has several times referred to the contrast between the implication of the IS-LM model that expansionary monetary policy implies a reduced interest rate, and Friedman’s oft-repeated dictum that high interest rates are a sign of easy money, and low interest rates a sign of tight money. This was a very clever strategic and rhetorical move by Scott, because it did highlight a key difference between Keynesian and Monetarist ideas while distracting attention from the overlap between Friedman and Keynesians on the basic analytics of nominal-income determination.

Alghough I agree with Scott that Friedman’s dictum that high interest rates distinguishes him from Keynes and Keynesian economists, I think that Scott leaves out an important detail: Friedman’s dictum also distinguishes him from just about all pre-Keynesian monetary economists. Keynes did not invent the terms “dear money” and “cheap money.” Those terms were around for over a century before Keynes came on the scene, so Keynes and the Keynesians were merely reflecting the common understanding of all (or nearly all) economists that high interest rates were a sign of “dear” or “tight” money, and low interest rates a sign of “cheap” or “easy” money. For example, in his magisterial A Century of Bank Rate, Hawtrey actually provided numerical bounds on what constituted cheap or dear money in the period he examined, from 1844 to 1938. Cheap money corresponded to a bank rate less than 3.5% and dear money to a bank rate over 4.5%, 3.5 to 4.5% being the intermediate range.

Take the period just leading up to the Great Depression, when Britain returned to the gold standard in 1925. The Bank of England kept its bank rate over 5% almost continuously until well into 1930. Meanwhile the discount rate of the Federal Reserve System from 1925 to late 1928 was between 3.5 and 5%, the increase in the discount rate in 1928 to 5% representing a decisive shift toward tight money that helped drive the world economy into the Great Depression. We all know – and certainly no one better than Scott – that, in the late 1920s, the bank rate was an absolutely reliable indicator of the stance of monetary policy. So what are we to make of Friedman’s dictum?

I think that the key point is that traditional notions of central banking – the idea of “cheap” or “dear” money – were arrived at during the nineteenth century when almost all central banks were operating either in terms of a convertible (gold or silver or bimetallic) standard or with reference to such a standard, so that the effect of monetary policy on prices could be monitored by observing the discount of the currency relative to gold or silver. In other words, there was an international price level in terms of gold (or silver), and the price level of every country could be observed by looking at the relationship of its currency to gold (or silver). As long as convertibility was maintained between a currency and gold (or silver), the price level in terms of that currency was fixed.

If a central bank changed its bank rate, as long as convertibility was maintained (and obviously most changes in bank rate occurred with no change in convertibility), the effect of the change in bank rate was not reflected in the country’s price level (which was determined by convertibility). So what was the point of a change in bank rate under those circumstances? Simply for the central bank to increase or decrease its holding of reserves (usually gold or silver). By increasing bank rate, the central bank would accumulate additional reserves, and, by decreasing bank rate, it would reduce its reserves. A “dear money” policy was the means by which a central bank could add to its reserve and an “easy money” policy was the means by which it could disgorge reserves.

So the idea that a central bank operating under a convertible standard could control its price level was based on a misapprehension — a widely held misapprehension to be sure — but still a mistaken application of the naive quantity theory of money to a convertible monetary standard. Nevertheless, although the irrelevance of bank rate to the domestic price level was not always properly understood in the nineteenth century – economists associated with the Currency School were especially confused on this point — the practical association between interest rates and the stance of monetary policy was well understood, which is why all monetary theorists in the nineteenth and early twentieth centuries agreed that high interest rates were a sign of dear money and low interest rates a sign of cheap money. Keynes and the Keynesians were simply reflecting the conventional wisdom.

Now after World War II, when convertibility was no longer a real constraint on the price level (despite the sham convertibility of the Bretton Woods system), it was a true innovation of Friedman to point out that the old association between dear (cheap) money and high (low) interest rates was no longer a reliable indicator of the stance of monetary policy. However, as a knee-jerk follower of the Currency School – the 3% rule being Friedman’s attempt to adapt the Bank Charter Act of 1844 to a fiat currency, and with equally (and predictably) lousy results – Friedman never understood that under the gold standard, it is the price level which is fixed and the money supply that is endogenously determined, which is why much of the Monetary History, especially the part about the Great Depression (not, as Friedman called it, “Contraction,” erroneously implying that the change in the quantity of money was the cause, rather than the effect, of the deflation that characterized the Great Depression) is fundamentally misguided owing to its comprehensive misunderstanding of the monetary adjustment mechanism under a convertible standard.

PS This is written in haste, so there may be some errors insofar as I relying on my memory without checking my sources. I am sure that readers will correct my lapses of memory

PPS I also apologize for not responding to recent comments, I will try to rectify that transgression over the next few days.

Sumner Sticks with Friedman

Scott Sumner won’t let go. Scott had another post today trying to show that the Cambridge Theory of the demand for money was already in place before Keynes arrived on the scene. He quotes from Hicks’s classic article “Mr. Keynes and the Classics” to dispute the quotation from another classic article by Hicks, “A Suggestions for Simplifying the Theory of Money,” which I presented in a post last week, demonstrating that Hicks credited Keynes with an important contribution to the demand for money that went beyond what Pigou, and even Lavington, had provided in their discussions of the demand for money.

In this battle of dueling quotations, I will now call upon Mark Blaug, perhaps the greatest historian of economics since Schumpeter, who in his book Economic Theory in Retrospect devotes an entire chapter (15) to the neoclassical theory of money, interest and prices. I quote from pp. 636-37 (4th edition).

Marshall and his followers went some way to move the theory of the demand for money in the direction of ordinary demand analysis, first, by relating money to net output or national income rather than the broader category of total transactions, and, second, by shifting from money’s rate of turnover to the proportion of annual income that the public wishes to hold in the form of money. In purely formal terms, there I nothing to choose between the Fisherian transaction approach and the Cambridge cash-balance approach, but the Cambridge formulation held out the potential of a genuine portfolio theory of the demand for money, which potential, however, was never fully exploited. . . .

The Cambridge formulation implies a demand for money equation, D_m = kPY, which contains no variable to represent the opportunity costs of holding cash, namely the rate of interest or the yield of alternative non-money assets, analogous to the relative price arguments of ordinary demand functions.
Yet a straight-forward application of utility-maximizing principles would have suggested that a rise in interest rates is likely to induce a fall in k as people strive to substitute interest-earning assets for passive money balances in their asset portfolios. Similarly, a fall in interest rates, by lowering the opportunity cost of holding money, is likely to cause a rise in k. Strangely enough, however, the Cambridge monetary theory never explicitly recognized the functional dependence of k on either the rate of interest or the rate on all non-monetary assets. After constructing a framework highly suggestive of a study of all the factors influencing cash-holding decisions, the Cambridge writers tended to lapse back to a list of the determinants of k that differed in no important respects from the list of institutional factors that Fisher had cited in his discussion of V. One can find references in Marshall, Pigou and particularly Lavington to a representative individual striking a balance between the costs of cash holdings in terms of interest foregone (minus the brokerage costs that would have been incurred by the movement into stocks and bonds) and their returns in terms of convenience and security against default but such passages were never systematically integrated with the cash-balance equation. As late as 1923, we find the young Keynes in A Tract on Monetary Reform interpreting k as a stable constant, representing an invariant link in the transmission mechanism connecting money to prices. If only Keynes at that date had read Wicksell instead of Marshall, he might have arrived at a money demand function that incorporates variations in the interest rate years before The General Theory (1936).

Moving to pp. 645-46, we find the following under the heading “The Demand for Money after Keynes.”

In giving explicit consideration to the yields on assets that compete with money, Keynes became one of the founders of the portfolio balance approach to monetary analysis. However, it is Hicks rather than Keynes who ought to be regarded as the founder of the view that the demand for money is simply an aspect of the problem of choosing an optimum portfolio of assets. In a remarkable paper published a year before the appearance of the General Theory, modestly entitled “A Suggestion for Simplifying the Theory of Money,” Hicks argued that money held at least partly as a store of value must be considered a type of capital asset. Hence the demand for money equation must include total wealth and expected rates of return on non-monetary assets as explanatory variables. Because individuals can choose to hold their entire wealth portfolios in the form of cash, the wealth variable represents the budget constraint on money holdings. The yield variables, on the other hand, represent both the opportunity costs of holding money and the substitutions effects of changes in relative rates of return. Individuals optimize their portfolio balances by comparing these yields with the imputed yield in terms of convenience and security of holding money. By these means, Hicks in effect treated the demand for money as a problem of balance sheet equilibrium analyzed along the same lines as those employed in ordinary demand theory.

It was Milton Friedman who carried this Hicksian analysis of money as a capital asset to its logical conclusion. In a 1956 essay, he set out a precise and complete specification of the relevant constraints and opportunity cost variable entering a household’s money demand function. His independent variable included wealth or permanent income – the present value of expected future receipts from all sources, whether personal earning or the income from real property and financial assets – the ratio of human to non-human wealth, expected rates of return on stocks, bonds and real assets, the nominal interest rate, the actual price level, and, finally, the expected percentage change in the price level. Like Hicks, Friedman specified wealth as the appropriate budget constraint but his concept of wealth was much broader than that adopted by Hicks. Whereas Keynes had viewed bonds as the only asset competing with cash, Friedman regarded all types of wealth as potential substitutes for cash holdings in an individual’s balance sheet; thus, instead of a single interest variable in the Keynesian liquidity preference equation, we get a whole list of relative yield variables in Friedman. An additional novel feature, entirely original with Friedman, is the inclusion of the expected rate of change in P as a measure of the anticipated rate of depreciation in the purchasing power of cash balances.

This formulation of the money demand function was offered in a paper entitled “The Quantity Theory of Money: A Restatement.” Friedman claimed not only that the quantity theory of money had always been a theory about the demand for money but also that his reformulation corresponded closely to what some of the great Chicago monetary economists, such as H.C. Simons and L. W. Mints, had always meant by the quantity theory. It is clear, however, from our earlier discussion that the quantity theory of money, while embodying an implicit conception of the demand for money, had always stood first and foremost for a theory of the determination of prices and nominal income; it contained much more than a particular theory of the demand for money.

Finally, Blaug remarks in his “notes for further reading” at the end of chapter 15,

In an influential essay, “The Quantity Theory of Money – A Restatement,” . . . M. Friedman claimed that his restatement was nothing more than the University of Chicago “oral” tradition. That claim was effectively destroyed by D. Patinkin, “The Chicago Tradition, the Quantity Theory, and Friedman, JMCB, 1969 .

Well, just a couple of quick comments on Blaug. I don’t entirely agree with everything he says about Cambridge monetary theory, and about the relative importance of Hicks and Keynes in advancing the theory of the demand for money. Cambridge economists may have been a bit more aware that the demand for money was a function of the rate of interest than he admits, and I think Keynes in chapter 17, definitely formulated a theory of the demand for money in a portfolio balance context, so I think that Friedman was indebted to both Hicks and Keynes for his theory of the demand for money.

As for Scott Sumner’s quotation from Hicks’s Mr. Keynes and the Classics, I think the point of that paper was not so much the theory of the demand for money, which had already been addressed in the 1935 paper from which I quoted, as to sketch out a way of generalizing the argument of the General Theory to encompass both the liquidity trap and the non-liquidity trap cases within a single graph. From the standpoint of the IS-LM diagram, the distinctive Keynesian contribution was the case of absolute liquidity preference, that doesn’t mean that Hicks meant that nothing had been added to the theory of the demand for money since Lavington. If that were the case, Hicks would have been denying that his 1935 paper had made any contribution. I don’t think that’s what he meant to suggest.

To sum up: 1) there was no Chicago oral tradition of the demand for money; 2) Friedman’s restatement of the quantity theory owed more to Keynes (and Hicks) than he admitted; 3) Friedman adapted the Cambridge/Keynes/Hicks theory of the demand for money in novel ways that allowed him to develop an analysis of price level changes that was more straightforward than was possible in the IS-LM model, thereby de-emphasizing the link between money and interest rates, which had been a such a prominent feature of the Keynesian models. That of course is a point that Scott Sumner likes to stress. In an upcoming post, I will comment on the fact that it was not just Keynesian models which stressed the link between money and interest rates. Pre-Keynesian monetary models also stressed the connection between easy money and low interest rates and dear money and high interest rates. Friedman’s argument was thus an innovation not only relative to Keynesian models but to orthodox monetary models. What accounts for this innovation?

My Milton Friedman Problem

In my previous post , I discussed Keynes’s perplexing and problematic criticism of the Fisher equation in chapter 11 of the General Theory, perplexing because it is difficult to understand what Keynes is trying to say in the passage, and problematic because it is not only inconsistent with Keynes’s reasoning in earlier writings in which he essentially reproduced Fisher’s argument, it is also inconsistent with Keynes’s reasoning in chapter 17 of the General Theory in his exposition of own rates of interest and their equilibrium relationship. Scott Sumner honored me with a whole post on his blog which he entitled “Glasner on Keynes and the Fisher Effect,” quite a nice little ego boost.

After paraphrasing some of what I had written in his own terminology, Scott quoted me in responding to a dismissive comment that Krugman recently made about Milton Friedman, of whom Scott tends to be highly protective. Here’s the passage I am referring to.

PPS.  Paul Krugman recently wrote the following:

Just stabilize the money supply, declared Milton Friedman, and we don’t need any of this Keynesian stuff (even though Friedman, when pressured into providing an underlying framework, basically acknowledged that he believed in IS-LM).

Actually Friedman hated IS-LM.  I don’t doubt that one could write down a set of equilibria in the money market and goods market, as a function of interest rates and real output, for almost any model.  But does this sound like a guy who “believed in” the IS-LM model as a useful way of thinking about macro policy?

Low interest rates are generally a sign that money has been tight, as in Japan; high interest rates, that money has been easy.

It turns out that IS-LM curves will look very different if one moves away from the interest rate transmission mechanism of the Keynesians.  Again, here’s David:

Before closing, I will just make two side comments. First, my interpretation of Keynes’s take on the Fisher equation is similar to that of Allin Cottrell in his 1994 paper “Keynes and the Keynesians on the Fisher Effect.” Second, I would point out that the Keynesian analysis violates the standard neoclassical assumption that, in a two-factor production function, the factors are complementary, which implies that an increase in employment raises the MEC schedule. The IS curve is not downward-sloping, but upward sloping. This is point, as I have explained previously (here and here), was made a long time ago by Earl Thompson, and it has been made recently by Nick Rowe and Miles Kimball.I hope in a future post to work out in more detail the relationship between the Keynesian and the Fisherian analyses of real and nominal interest rates.

Please do.  Krugman reads Glasner’s blog, and if David keeps posting on this stuff then Krugman will eventually realize that hearing a few wisecracks from older Keynesians about various non-Keynesian traditions doesn’t make one an expert on the history of monetary thought.

I wrote a comment on Scott’s blog responding to this post in which, after thanking him for mentioning me in the same breath as Keynes and Fisher, I observed that I didn’t find Krugman’s characterization of Friedman as someone who basically believed in IS-LM as being in any way implausible.

Then, about Friedman, I don’t think he believed in IS-LM, but it’s not as if he had an alternative macromodel. He didn’t have a macromodel, so he was stuck with something like an IS-LM model by default, as was made painfully clear by his attempt to spell out his framework for monetary analysis in the early 1970s. Basically he just tinkered with the IS-LM to allow the price level to be determined, rather than leaving it undetermined as in the original Hicksian formulation. Of course in his policy analysis and historical work he was not constained by any formal macromodel, so he followed his instincts which were often reliable, but sometimes not so.

So I am afraid that my take may on Friedman may be a little closer to Krugman’s than to yours. But the real point is that IS-LM is just a framework that can be adjusted to suit the purposes of the modeler. For Friedman the important thing was to deny that that there is a liquidity trap, and introduce an explicit money-supply-money-demand relation to determine the absolute price level. It’s not just Krugman who says that, it’s also Don Patinkin and Harry Johnson. Whether Krugman knows the history of thought, I don’t know, but surely Patinkin and Johnson did.

Scott responded:

I’m afraid I strongly disagree regarding Friedman. The IS-LM “model” is much more than just the IS-LM graph, or even an assumption about the interest elasticity of money demand. For instance, suppose a shift in LM also causes IS to shift. Is that still the IS-LM model? If so, then I’d say it should be called the “IS-LM tautology” as literally anything would be possible.

When I read Friedman’s work it comes across as a sort of sustained assault on IS-LM type thinking.

To which I replied:

I think that if you look at Friedman’s responses to his critics the volume Milton Friedman’s Monetary Framework: A Debate with his Critics, he said explicitly that he didn’t think that the main differences among Keynesians and Monetarists were about theory, but about empirical estimates of the relevant elasticities. So I think that in this argument Friedman’s on my side.

And finally Scott:

This would probably be easier if you provided some examples of monetary ideas that are in conflict with IS-LM. Or indeed any ideas that are in conflict with IS-LM. I worry that people are interpreting IS-LM too broadly.

For instance, do Keynesians “believe” in MV=PY? Obviously yes. Do they think it’s useful? No.

Everyone agrees there are a set of points where the money market is in equilibrium. People don’t agree on whether easy money raises interest rates or lowers interest rates. In my view the term “believing in IS-LM” implies a belief that easy money lowers rates, which boosts investment, which boosts RGDP. (At least when not at the zero bound.) Friedman may agree that easy money boosts RGDP, but may not agree on the transmission mechanism.

People used IS-LM to argue against the Friedman and Schwartz view that tight money caused the Depression. They’d say; “How could tight money have caused the Depression? Interest rates fell sharply in 1930?”

I think that Friedman meant that economists agreed on some of the theoretical building blocks of IS-LM, but not on how the entire picture fit together.

Oddly, your critique of Keynes reminds me a lot of Friedman’s critiques of Keynes.

Actually, this was not the first time that I provoked a negative response by writing critically about Friedman. Almost a year and a half ago, I wrote a post (“Was Milton Friedman a Closet Keynesian?”) which drew some critical comments from such reliably supportive commenters as Marcus Nunes, W. Peden, and Luis Arroyo. I guess Scott must have been otherwise occupied, because I didn’t hear a word from him. Here’s what I said:

Commenting on a supremely silly and embarrassingly uninformed (no, Ms. Shlaes, A Monetary History of the United States was not Friedman’s first great work, Essays in Positive Economics, Studies in the Quantity Theory of Money, A Theory of the Consumption Function, A Program for Monetary Stability, and Capitalism and Freedom were all published before A Monetary History of the US was published) column by Amity Shlaes, accusing Ben Bernanke of betraying the teachings of Milton Friedman, teachings that Bernanke had once promised would guide the Fed for ever more, Paul Krugman turned the tables and accused Friedman of having been a crypto-Keynesian.

The truth, although nobody on the right will ever admit it, is that Friedman was basically a Keynesian — or, if you like, a Hicksian. His framework was just IS-LM coupled with an assertion that the LM curve was close enough to vertical — and money demand sufficiently stable — that steady growth in the money supply would do the job of economic stabilization. These were empirical propositions, not basic differences in analysis; and if they turn out to be wrong (as they have), monetarism dissolves back into Keynesianism.

Krugman is being unkind, but he is at least partly right.  In his famous introduction to Studies in the Quantity Theory of Money, which he called “The Quantity Theory of Money:  A Restatement,” Friedman gave the game away when he called the quantity theory of money a theory of the demand for money, an almost shockingly absurd characterization of what anyone had ever thought the quantity theory of money was.  At best one might have said that the quantity theory of money was a non-theory of the demand for money, but Friedman somehow got it into his head that he could get away with repackaging the Cambridge theory of the demand for money — the basis on which Keynes built his theory of liquidity preference — and calling that theory the quantity theory of money, while ascribing it not to Cambridge, but to a largely imaginary oral tradition at the University of Chicago.  Friedman was eventually called on this bit of scholarly legerdemain by his old friend from graduate school at Chicago Don Patinkin, and, subsequently, in an increasingly vitriolic series of essays and lectures by his then Chicago colleague Harry Johnson.  Friedman never repeated his references to the Chicago oral tradition in his later writings about the quantity theory. . . . But the simple fact is that Friedman was never able to set down a monetary or a macroeconomic model that wasn’t grounded in the conventional macroeconomics of his time.

As further evidence of Friedman’s very conventional theoretical conception of monetary theory, I could also cite Friedman’s famous (or, if you prefer, infamous) comment (often mistakenly attributed to Richard Nixon) “we are all Keynesians now” and the not so famous second half of the comment “and none of us are Keynesians anymore.” That was simply Friedman’s way of signaling his basic assent to the neoclassical synthesis which was built on the foundation of Hicksian IS-LM model augmented with a real balance effect and the assumption that prices and wages are sticky in the short run and flexible in the long run. So Friedman meant that we are all Keynesians now in the sense that the IS-LM model derived by Hicks from the General Theory was more or less universally accepted, but that none of us are Keynesians anymore in the sense that this framework was reconciled with the supposed neoclassical principle of the monetary neutrality of a unique full-employment equilibrium that can, in principle, be achieved by market forces, a principle that Keynes claimed to have disproved.

But to be fair, I should also observe that missing from Krugman’s take down of Friedman was any mention that in the original HIcksian IS-LM model, the price level was left undetermined, so that as late as 1970, most Keynesians were still in denial that inflation was a monetary phenomenon, arguing instead that inflation was essentially a cost-push phenomenon determined by the rate of increase in wages. Control of inflation was thus not primarily under the control of the central bank, but required some sort of “incomes policy” (wage-price guidelines, guideposts, controls or what have you) which opened the door for Nixon to cynically outflank his Democratic (Keynesian) opponents by coopting their proposals for price controls when he imposed a wage-price freeze (almost 42 years ago on August 15, 1971) to his everlasting shame and discredit.

Scott asked me to list some monetary ideas that I believe are in conflict with IS-LM. I have done so in my earlier posts (here, here, here and here) on Earl Thompson’s paper “A Reformulation of Macroeconomic Theory” (not that I am totally satisfied with Thompson’s model either, but that’s a topic for another post). Three of the main messages from Thompson’s work are that IS-LM mischaracterizes the monetary sector, because in a modern monetary economy the money supply is endogenous, not exogenous as Keynes and Friedman assumed. Second, the IS curve (or something corresponding to it) is not negatively sloped as Keynesians generally assume, but upward-sloping. I don’t think Friedman ever said a word about an upward-sloping IS curve. Third, the IS-LM model is essentially a one-period model which makes it difficult to carry out a dynamic analysis that incorporates expectations into that framework. Analysis of inflation, expectations, and the distinction between nominal and real interest rates requires a richer model than the HIcksian IS-LM apparatus. But Friedman didn’t scrap IS-LM, he expanded it to accommodate expectations, inflation, and the distinction between real and nominal interest rates.

Scott’s complaint about IS-LM seems to be that it implies that easy money reduces interest rates and that tight money raises rates, but, in reality, it’s the opposite. But I don’t think that you need a macro-model to understand that low inflation implies low interest rates and that high inflation implies high interest rates. There is nothing in IS-LM that contradicts that insight; it just requires augmenting the model with a term for expectations. But there’s nothing in the model that prevents you from seeing the distinction between real and nominal interest rates. Similarly, there is nothing in MV = PY that prevented Friedman from seeing that increasing the quantity of money by 3% a year was not likely to stabilize the economy. If you are committed to a particular result, you can always torture a model in such a way that the desired result can be deduced from it. Friedman did it to MV = PY to get his 3% rule; Keynesians (or some of them) did it to IS-LM to argue that low interest rates always indicate easy money (and it’s not only Keynesians who do that, as Scott knows only too well). So what? Those are examples of the universal tendency to forget that there is an identification problem. I blame the modeler, not the model.

OK, so why am I not a fan of Friedman’s? Here are some reasons. But before I list them, I will state for the record that he was a great economist, and deserved the professional accolades that he received in his long and amazingly productive career. I just don’t think that he was that great a monetary theorist, but his accomplishments far exceeded his contributions to monetary theory. The accomplishments mainly stemmed from his great understanding of price theory, and his skill in applying it to economic problems, and his great skill as a mathematical statistician.

1 His knowledge of the history of monetary theory was very inadequate. He had an inordinately high opinion of Lloyd Mints’s History of Banking Theory which was obsessed with proving that the real bills doctrine was a fallacy, uncritically adopting its pro-currency-school and anti-banking-school bias.

2 He covered up his lack of knowledge of the history of monetary theory by inventing a non-existent Chicago oral tradition and using it as a disguise for his repackaging the Cambridge theory of the demand for money and aspects of the Keynesian theory of liquidity preference as the quantity theory of money, while deliberately obfuscating the role of the interest rate as the opportunity cost of holding money.

3 His theory of international monetary adjustment was a naïve version of the Humean Price-Specie-Flow mechanism, ignoring the tendency of commodity arbitrage to equalize price levels under the gold standard even without gold shipments, thereby misinterpreting the significance of gold shipments under the gold standard.

4 In trying to find a respectable alternative to Keynesian theory, he completely ignored all pre-Keynesian monetary theories other than what he regarded as the discredited Austrian theory, overlooking or suppressing the fact that Hawtrey and Cassel had 40 years before he published the Monetary History of the United States provided (before the fact) a monetary explanation for the Great Depression, which he claimed to have discovered. And in every important respect, Friedman’s explanation was inferior to and retrogression from Hawtrey and Cassel explanation.

5 For example, his theory provided no explanation for the beginning of the downturn in 1929, treating it as if it were simply routine business-cycle downturn, while ignoring the international dimensions, and especially the critical role played by the insane Bank of France.

6 His 3% rule was predicated on the implicit assumption that the demand for money (or velocity of circulation) is highly stable, a proposition for which there was, at best, weak empirical support. Moreover, it was completely at variance with experience during the nineteenth century when the model for his 3% rule — Peel’s Bank Charter Act of 1844 — had to be suspended three times in the next 22 years as a result of financial crises largely induced, as Walter Bagehot explained, by the restriction on creation of banknotes imposed by the Bank Charter Act. However, despite its obvious shortcomings, the 3% rule did serve as an ideological shield with which Friedman could defend his libertarian credentials against criticism for his opposition to the gold standard (so beloved of libertarians) and to free banking (the theory of which Friedman did not comprehend until late in his career).

7 Despite his professed libertarianism, he was an intellectual bully who abused underlings (students and junior professors) who dared to disagree with him, as documented in Perry Mehrling’s biography of Fischer Black, and confirmed to me by others who attended his lectures. Black was made so uncomfortable by Friedman that Black fled Chicago to seek refuge among the Keynesians at MIT.


About Me

David Glasner
Washington, DC

I am an economist at the Federal Trade Commission. Nothing that you read on this blog necessarily reflects the views of the FTC or the individual commissioners. Although I work at the FTC as an antitrust economist, most of my research and writing has been on monetary economics and policy and the history of monetary theory. In my book Free Banking and Monetary Reform, I argued for a non-Monetarist non-Keynesian approach to monetary policy, based on a theory of a competitive supply of money. Over the years, I have become increasingly impressed by the similarities between my approach and that of R. G. Hawtrey and hope to bring Hawtrey's unduly neglected contributions to the attention of a wider audience.

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 266 other followers


Follow

Get every new post delivered to your Inbox.

Join 266 other followers