The Costs of a Gold Standard

0:00Well, the topic I have for this afternoon requires us to shift gears, I think, shifting from history to some theoretical issues, shifting from concern with the pre-Civil War period to the modern period. The issue of the cost of the gold standard has been an issue that's interested me for some time, and I think maybe because of the understanding of this issue on the part of the lay person. Thank you. I'm a little taller than Murray. The understanding of this issue on the part of the lay person tends to be about the same as the understanding on the part of professional economists, so it seems to me, and we may want to be optimistic and think that somehow the lay person has raised himself to the level of understanding of the professional economist, But I'm afraid that it tends to go in the other direction here.

1:01The economist viewpoint doesn't seem to be much better than the man in the streets viewpoint on this particular issue, the cost of the gold standard. Costs generally are simply perceived as being high, prohibitively high, and this is one reason, a primary reason, in some people's judgment for rejecting the gold standard. Although the paper itself is on the cost of a gold standard, the reader would be disappointed if he were to thumb through the paper looking for all sorts of cost accounting or lists of costs, charts or tables or something of this sort, figures that we could all add up and find out what after all the costs really are.

1:48Rather, the focus on cost in my paper has simply allowed me to grapple with the basic issues of the gold standard in what I think to be fairly helpful and illuminating way. I might mention that the thing that sort of piqued my interest in the gold standard and the cost issue in particular in recent times is that I attended another conference on monetary stability earlier this year. And I heard Alan Meltzer mention that the cost of the gold standard amounted to about 16% of the economy's real growth in a given year. And from his own paper and from his comments, it was not at all clear where this number came from, how it was derived.

2:39and so I thought it might be interesting to look into that, more about that later on in my talk this afternoon. I think first what we need to do is get some preliminaries out of the way and make it clear about what sort of a gold standard that we're talking about when we talk about the cost of the gold standard. I really only insist on two particular characteristics that the gold standard have to make my discussion relevant. One would be that gold really is money in some meaningful sense. This allows us to make the distinction between what Milton Friedman has called a real gold standard, where gold serves as money, and what he calls a pseudo gold standard, which really is better described as a price support program for the gold industry.

3:33I might mention here that Friedman in the Chicago School, even though Friedman himself made this distinction very clear, it seemed to me in modern debate to always assume that the gold standard being advocated And I think that any intelligent debate requires that this misunderstanding be cleared up before discussing further issues. The second criteria that the gold standard has to meet to make my comments relevant is that the gold standard has to be a decentralized system. In other words, the gold has to be provided privately. privately, it's not managed by a central bank or the treasury or any other central agency of the government.

4:22So it's truly a free market gold system in which gold does serve as money. Well, we might spend a minute or two simply considering how such a system would work. It really is fairly simple. It only requires that we understand some basic microeconomic principles. But in the event that there is an increase in the demand for money, demand for gold, for whatever reason, that sets into motion market processes that will respond to that change in demand. In other words, there will be downward pressure on prices in general, which is to say that there will be upward pressure on the value of gold, gold, upward pressure on the prices of factors of production that produce gold, resources will be allocated in part away from the production of goods and services and into the production of gold, which by hypothesis is in greater demand, and the provision of more gold will relieve the downward pressure on price to some extent, and to some extent prices will will actually fall to accommodate the adjustment in the increase in the demand for gold.

5:40Now, when I began reading into the literature to try to figure out just what is wrong with this system in the eyes of the orthodox economists, the Keynesians, let's say, and the monetarists, I realized I had a problem that I had not been aware of before, And that is that the two groups had different views on the amount of resources allocated to gold. The Keynesians, in fact, and here I'm using Keynes himself, the general theory is the authority on what Keynes had to say and not his interpreters, the Keynesians didn't like gold because it didn't use enough resources. And they reasoned something as follows, that when there's downward pressure on price, they assume prices didn't adjust at all.

6:30And that all adjustments had to be made in terms of quantities, that labor and capital had to be shifted out of the production of goods and services and into the production of gold. And unfortunately, it just didn't take that many laborers, it didn't take that much capital to dig out the gold that would satisfy the demand and for that reason they found gold to be an unsatisfactory monetary commodity. So that I'm not accused of misinterpreting Keynes, I would like to read one statement from The General Theory and he says that it is interesting to notice that the characteristic which has been traditionally supposed to render gold especially suitable for use as a standard of Value, namely its inelasticity of supply, turns out to be precisely the characteristic which is at the bottom of the trouble. In other words, he wanted to see a commodity whose supply would be much more elastic so that much more labor and much more resources could be used in mining the monetary commodity. And I can't resist also giving you the flavor of Keynes

7:40in this particular passage so that you can see what sort of a vision he really had Unemployment develops, that is to say, because people want the moon. Men cannot be employed when the object of their desire, that is money, is something which cannot be produced and the demand for which cannot readily be choked off. Keynes goes on to say, there is no remedy, but to persuade the public that green cheese is practically the same thing and to have a green cheese factory, i.e. a central bank, under public control. This was the Keynesian solution to the problem, and as he saw the problem, gold production didn't use enough resources.

8:35The monetarist view turns out to be just the opposite, that the monetarists tend to take any resources that the gold industry uses at all as being too many resources. The monetarists would like to see either an immediate and painless adjustment of price to accommodate any increase in demand, or alternatively, They would like to see some fiat standard in which money could be costlessly produced to meet whatever demand there is in such a way that no resources, whatever, would have to be devoted to the production of gold. Now, the Austrian economist, Misesians, Hayekians, would be ill-advised, I think, to try to argue, either on Keynesian grounds or on monetarist grounds, that the amount of gold actually being mined or that the amount of resources being committed to that purpose are just enough but not too much.

9:37I'm not going to talk too much, okay, because whatever the quantity is that's being allocated to gold mining is going to be too many resources in the view of the monetarists and not enough resources in the view of the Keynesians. And some politicians, it seems, tend not to hesitate to use both of those arguments at the same time. Let me move on to actual estimates that are made about the resource costs of gold and this will allow us to see the significance of Alan Meltzer's 16% figure. It turns out that Meltzer was quoting an article by Friedman written back in the early 50s.

10:23He didn't indicate to us how Friedman made his calculations, but he simply updated Friedman's calculations in terms of the current ratio of income to money. So I went back and looked at the Friedman estimate, and what I discovered was that Friedman simply assumed that the way to estimate the cost of the gold standard is to assume that whatever demand for money there is will be fully met by the production of gold, that it will never be met by declining prices of goods and services. In other words, he simply assumed that whatever the real rate of growth is in the economy, It would have to be matched by an equal rate of the production of monetary gold.

11:19Then he observed that the total quantity of gold in the economy is about half of the total GNP, which means that if the gross national product were to increase at a rate of, say, 3%, then money supply would have to increase at that same rate. And since money constitutes half of GNP, then 1.5% of the economy's total GNP would have to be devoted to the production of gold. He adjusted these figures for changes in velocity, which is another way of measuring changes in the demand for money apart from the growth in the economy.

12:04and he found that the cost figures are a little higher still. Now, I found this a peculiar way to go about estimating the cost of gold for a number of reasons. One is that it has implications that the monitors tend to gloss over, and that is that their estimate is tied very closely to the growth rate of the economy. And what that implies, for instance, is that in recent periods, when the growth rate of the economy has been zero, that would imply that the resource costs of gold are zero, based on their own calculations. You get a very ironic result in that it's costless to re-adopt the gold standard when the economy is not growing, especially ironic when we realize that the reason the economy is not growing is because it's suffering under unsound money in the current period.

13:03So that the monitors may see the cost now as zero, we can go back on the gold standard costlessly, but by their own calculations, as soon as the economy begins to grow again because of its adherence to sound money, the gold standard would once again become too costly and would have to be abandoned, I suppose. I haven't, I've never heard of monitors address this particular point Now, we can get some insights into these estimates if we see what this implies about the supply of gold. Well, to assume that enough gold gets mined to keep the price level constant is to assume that the supply curve of gold or the supply of gold is perfectly elastic, okay?

13:58And yet the older economists always claim that one of the primary benefits of gold is its inelasticity, that it's a scarce commodity, that not much of it is yielded each year, and that its scarcity is what made it a particularly suitable commodity. But these estimates assume that the elasticity is infinite, that it's a perfectly elastic in supply, okay? So really what the monetarist has estimated is not the resource costs of gold, but the resource costs of adopting a perfectly elastic commodity as a monetary standard.

14:44I suppose the only, one of the few actual standards proposed of that sort is the one by C.O. Hardy a number of decades ago where he proposed He says that the common brick be used as a monetary standard and he saw this in very Keynesian terms. In other words, if workers, I think he discovered that something like 96% of all the counties in the United States had clay that was suitable for producing bricks. So this would be a good solution for the economy's ills. the workers became unemployed, well they could simply go home and begin making money in their backyards and once they molded the bricks and spent that, that would stimulate the economy and they'd get their jobs back.

15:30So this is the kind of standard that seems to be estimated by Friedman and again by Melter earlier this year. Now, I think there's something even more fundamental that underlies this willingness to base their estimates on a perfectly elastic supply of money. And that is that they take the ultimate goal or the policy objective of the monetary authority to be the objective of maintaining a stable price level. All right, and I think it's worth our while to consider this issue for a few minutes and to see what the history of this is. Why is it that the constant price level is seen as the appropriate target, who simply assumed that that was the objective?

16:24Cassel said it seems to be the simplest assumption to make. Begou was sure that if the government pursued policies that would maintain stable prices, monetary factors would not interfere with the workings of the market. American economists tend to adopt the same assumption. Clark Warburton is one I cite, and Milton Friedman followed suit, adopting the same assumption, Friedman admitted very clearly that it was simply an assumption, an assumption, Friedman said, made for the sake of convenience. And I would think that this would be disquieting for the opponents of gold to realize that the estimates of resource costs that they use to argue against the gold standard are based on an assumption that was made simply for convenience.

17:13Friedman does, in one of his classic articles, The Optimum Quantity of Money, decide to question this assumption and see if it's not possible to theoretically derive an optimum change in the price level. And he manages to employ the calculus of marginalism to equate marginal cost to marginal benefits and solve for the rate of decline of the money supply. He came up with some conclusions that indicated that people would be better off if the price level were actually falling, okay?

17:58and he reasoned that people's real cash balances would be higher under a falling price level and yet no other real factors in his calculations at least were assumed to change. Therefore he concluded that the optimum price deflation was something like four or five percent. Turns out he failed even to convince himself of this argument. He ends the article with what he called a final schizophrenic note in which he sort of teetered between recommending falling prices and recommending constant prices. And since that article, which appeared I think in 1969, the economic profession has pretty much gone back to its old assumption of constant prices.

18:45Now I think here we can put the monetarists, the Keynesians and the Austrians in perspective that's helpful. The Keynesians judge monetary policy on the basis of its ability to create full employment. Stable money means money that will create full employment. For the monetarists, stable money means money that will give rise to constant prices. And for the Austrians, of course, stable money just means money that's chosen by the market participants themselves and not a money that's foisted on them by some central monetary manager. Now, putting it in perspective this way, I wouldn't want to suggest that full employment is per se undesirable, it's certainly not, or that constant prices per se are undesirable, it's certainly not, but they're not appropriate policies for government.

19:44In fact, what we discover in the case of both the Keynesians and the monitors is that when the government aims at full employment, it tends to generate a lot of unemployment and when it aims, at least in its rhetoric, at stable prices, it tends to generate a lot of price instability, certainly more unemployment, more instability of prices in comparison with the performance we would get under gold standard. Well, I don't really know of counter arguments to some of these claims. I don't see the monitors dealing with the Austrian arguments. They may express dissatisfaction with the arguments, but I don't see any attempt to rebut point by point some of these arguments of the Austrian school.

20:41I do see a couple of reasons for people clinging to the idea that what we want to achieve through monetary policy is stable prices. One is a political feasibility argument. It's believed that it might be easier to get the central bank to behave itself, that is, to limit the printing of money, than it would be to get it to disband altogether. I don't have to agree it would be easier, but I think that advocates of the gold standard don't expect the Federal Reserve Bank to begin behaving any time soon, and they take that as one of the reasons to advocate dismantling of the Fed rather than efforts to try to get it to behave.

21:28A second reason that people tend to favor stable money is because they tend to think of money as being simply a measure, a unit of account, a unit that is analogous to units of weight and units of length. And these arguments have some plausibility. They catch our attention, I think. Of course, carpenters would not fare well in their trade if they had measuring tools that fluctuated on their own. I point out in my paper truck drivers would dread weigh stations even more than they do already if they had to wonder how heavy a pound is today. Okay? We have to have stable units. We have to have units of weight and units of length that don't change.

22:20And by analogy, so too, we have to have a stable monetary unit, one that doesn't change from time to time. Well, this analogy has a certain appeal to it, and it may be just the right medicine to argue against, say, advocates of inflation or to argue against people who advocate cheap credit, even though that will eventually lead to inflation. But I argue that for advocates of sound money, we have more to learn from the sense in which that analogy doesn't hold than the sense in which it does. It's possible to have a standard unit of length and a standard unit of weight, but there simply is no...

23:06There is no immutable invariant unit of value. That's simply not the nature of value. All values are relative and they change over time. The attempt to search out a value that never changes is a throwback, I argue, to the old pre-subjectivist, pre-marginal economists. And to keep searching for such a thing is simply In my paper, I offer a counter analogy, an analogy that tries to capture the essence of a unit of value more closely than the unit of weight and length. And I argue that the monetary commodity really is more of a reference commodity, a benchmark.

23:53And I argue that the monetary commodity really is more of a reference commodity, a benchmark, a base point, okay? And if it has a physical analog, it would be something like an immutable reference point in the cosmos, let's say, a point that is forever fixed with respect to all the heavenly bodies. And we can imagine people submitting proposals on how to calculate where this reference point is. It may be indexation schemes to take into account the relative location of all of the heavenly bodies at any given point in time. But such an exercise would be as elusive as it is useless.

24:38After all, the most relevant point for our own purposes is the point where cosmic developments have left us. And so it is with value, I argue, that the most relevant value for purposes of measuring value is the monetary commodity that economic developments have given us. And if economic developments have gravitated towards the gold standard and gold is taken as the benchmark, the reference value, then that's the one that becomes relevant. whether or not it falls or rises in comparison with some calculated index or some price level conceived by somebody. Well, now let me return to the resource cost issue once more and point out what I see as two deficiencies in their argument that pretty much does away with the argument.

25:37One is that those cost calculations are much too narrowly conceived. Resource costs, after all, are only part of the costs of any standard, whether we're talking about a gold standard or paper standard. And if we limit our comparisons to resource costs, we're really not comparing the relevant costs. This is particularly true in the case of comparing institutions, alternative institutional arrangements. As an example, I mentioned that we might say that the resource costs of a penal institution that segregated criminals from the rest of society may be higher than the resource costs of an institution that just slapped the criminal's hand and turned the back loose in society. Well, that may be so. The resource costs may be higher, but the total costs, which would have to include the future crimes perpetrated by these criminals, would certainly not be higher.

26:29The resource costs here are just not relevant. The same problem occurs with monetary systems. If we want to consider the total costs of both gold and paper, we have to look at some critical costs of the paper standard that sometimes are overlooked. Cost of a paper standard includes the cost imposed on society of different political factions trying to get control of the printing press, the cost imposed by special interest Special Interest Groups, trying to get the printing press to cater to them, printing money, benefiting the special interest groups, cost imposed by paper money issuers in terms of disrupting economic activity through their inflation as they cave into these special interest groups, costs incurred by businessmen in their attempt to hedge against future inflation caused by the monetary. All these have to count as costs too, okay? Just because they're not, quote, resource costs doesn't mean This really captures the meaning of a seminal leave that I've heard attributed to Alan Greenspan, one that I wish the press would pick up on and ask Greenspan about it.

27:40Greenspan said at one time that putting the state in control of paper money is like putting a penny in the fuse box. And it may be true that the resource costs of the penny are less than the resource costs of the fuse, but surely the total costs which have to take into account the possibility of a disastrous fire would far outweigh the fuse. Now, people who are opponents of the gold standard tend to overlook the costs caused by the paper standard but to itemize costs over and above resource costs that we get under a gold standard.

28:25And this usually amounts to costs in terms of the time-consuming adjustments of prices that have to be undertaken when increases in monetary demand are not fully met with increases in the quantity of gold. In other words, if there's an increase in the demand for money, some prices have to fall in order to make the money in existence satisfy the existing demand. But I think there are critical problems with this view. Usually it's conceived that we have productivity growth, the economy is growing, and therefore, if you're under a gold standard, prices have to fall.

29:12The problem disappears as soon as we see that the increases in productivity are not themselves economy-wide increases. That is, we really get increases in the productivity of some processes, we get prices of some goods falling, we get prices of others falling. Increases in productivity are not themselves economy-wide increases, that is, we really get increases in the productivity of some processes, we get prices of some goods falling, we get prices of other goods rising, we get a change in the pattern of prices across the economy in response to productivity changes. Now, if you're on a gold standard, that just means that some prices that are falling will have to fall a little farther, some prices that are rising won't have to rise quite so much, some prices that would have risen won't rise at all.

30:07The prices just change in a different pattern, different from the one that would have governed had the productivity changes not taken place. Even if the government could stabilize some price level, it would not be easy to determine which price level to stabilize. Consider a growing economy in which the real rate of interest is falling. If the government stabilizes the consumer price index, well that means that the price of the factors has to rise a little bit to reflect the fallen interest rate. If the government stabilizes the factor prices, that means the consumer prices have to fall a little bit to accommodate the change in the interest rate.

30:54There really is no price level that could be stabilized that would prevent the prices on an economy-wide basis from adjusting to increases in productivity. Well, I go on in my paper to argue that the resource costs of gold are, what I call it, doubly irrelevant. And the first aspect of the irrelevancy is that it's a poor proxy for total costs, that we need to look at total costs and not resource costs. But the second argument, which I've seen made nowhere else in the literature, maybe it has been made, if some of you know about this, I'd like you to tell me. And that is that counting the resource costs of gold against the gold standard implies that if you had a paper standard, all these costs would be eliminated.

31:47In other words, people tend to see the costs of gold, mining it, storing it, mincing it into coins or bars, guarding it, all the rest. They see these as collectively wasteful activities, costly activities, and they count them against the gold standard and want a paper standard instead. But for this argument to hold up requires that these costs be eliminated when you go to a paper standard. Well, as soon as we make that assumption explicit, we see the fallacy in the argument. It really takes a naive view that the state can somehow repeal laws of economics to hold this view. The costs aren't eliminated at all. All under a paper standard, we still mine gold, we still min it into bars, we still store it, we still guard it.

32:34We incur all those resource costs that we would have incurred under the gold standard. And arguably, we incur more if the monetary managers mismanage the paper money system enough, If we have enough, we will actually drive up the price of gold and allocate more resources to gold mining than we would have under a gold standard. So even however high the costs of gold are in dollar terms, they're not costs that are relevant in the decision of whether to adopt a gold standard or a paper standard. I'll just turn in a minute or so I have remaining to talk about the benefit side of the lecture.

33:20I can't talk about costs the whole time without at least mentioning benefits. I think one of the reasons that the costs seem high to laypeople and even to economists is that they neglect to look at the benefits. If you look at the benefits of a gold standard, you tend to see that the costs are very small by comparison. As Mises has pointed out, the primary benefit of the gold standard is simply that it makes the monetary system immune from government tinkering. This, I think, is the only thing that we need to hold up as a major benefit. We don't want to claim that the government can never override the gold standard. and historically it's shown that it can, but it takes a long time, a long struggle, through use of coercion and propaganda to get the market to disband gold and during that time the economy can enjoy growth under sound money.

34:22Let me just conclude by saying that ultimately the cost of any action or any commodity, any institution, is the alternative action, commodity or institution foregone. The opportunity costs are the only cost that counts. The cost of one institution is foregoing another institution. The cost of a gold standard is foregoing a paper standard. The cost of sound money is foregoing unsound money.