A Century of Failure: Why It's Time to Consider Replacing the Fed

0:00I'm not going to talk very much about the future, but I do want to say some things about the past. I want to talk about the Federal Reserve's overall record. And before I do so, I do want to make a kind of apology about this talk, because I realize that an audience like this one probably doesn't need to be convinced that the Fed is a failure. But nevertheless, I hope that you'll find the talk interesting because what it actually is is a talk aimed at showing that the Fed is not just a failure in our terms but a failure in its own terms and in those of the mainstream economics profession and that this is so despite the fact that your average mainstream economist continues to have a very complacent The Federal Reserve's attitude about the Fed, and if pressed on the subject, will insist that the Fed has, after all, and despite the recent challenges it has faced, that the Fed has been a factor that has contributed to the well-being of the American economy, and that the statistics available will show that.

1:14What I want to do today is to show you that that position is in fact completely unsound and that it is unsound in terms of the best available statistics that these very mainstream economists have provided us with. So it's a kind of imminent criticism of the Fed that I intend to give you today and it is a criticism once again on its own terms and on the terms of those who are its apologists. I don't know if we can turn down the spots a little bit on the slides. Is that possible? Some of these slides will require... I think that's fine. Can you see those clearly? I want to start by reminding you about the background of the creation of the Federal Reserve System. Just over 100 years ago today, 1907, we had a very terrible financial crisis, The Panic of 1907 And it was approximately, at least, in response to this crisis that the Federal Reserve system was established.

2:21Now the Panic of 1907 was only one in a series of financial panics and crises that had beset the American economy in the years after the Civil War. It was also, of course, one of the worst. And there had been numerous attempts set reform of the monetary system up to the time of the panic of 1907, but that panic led to the establishment of something called the National Monetary Commission. And what the National Monetary Commission was, was a body set up to investigate both the nature and causes of the crises that had affected the U.S. economy up to that time, and alternatives to the existing monetary system with the aim of looking for a solution.

3:10Now I have to say it was a loaded commission. By the time this commission was set up it was very clear that the intentions of the people behind it, especially Nelson Aldrich who was behind the passage of the bill in question was that there should be an outcome favoring the establishment of a central bank. But what I want to convince you of is that today there is no less need to consider alternatives to the Fed than there was then and indeed the statistics suggest that we should be less happy with our monetary system now than people were in 1907. And we need something like a National Monetary Commission, though hopefully one that will will really try to establish an improved system.

4:03The outcome of the National Monetary System was, of course, not that disaster, but this one, the establishment of the Federal Reserve System. And it's that system, of course, that was supposed to provide for much greater stability and a much improved monetary arrangement than that which preceded it, than the national Here are some elements of the Fed's mission, as stated in its own publications or in the Federal Reserve Act itself, and they include, as you can see, promoting effectively the The Goals of Maximum Employment, Stable Prices, Moderate Long-Term Interest, and Moderate Long-Term Interest Rates.

4:53Elsewhere, the Fed says that it's responsible for containing financial disruptions and preventing their spread outside the financial sector. So these suggest, these goals, these official goals of the Fed suggest criteria by which which we should evaluate the system's success, and that's what I intend to do. I'm not going to say a whole lot about employment and unemployment, for the simple reason that as you all know, the factors that are responsible for the unemployment rate are, well, quite numerous. And the role of monetary policy in influencing unemployment is certainly important, are certainly important, but it's hard to separate the influence of monetary developments from other developments. Nevertheless, for what it's worth, we can ask, has the Federal Reserve era brought us an improved lower rate of employment than we saw before?

5:55And here's what the available statistics show. What I've done on these charts, I've tried to supply a vertical line indicating where the Fed was founded and other vertical lines to show significant cutoffs from other regime changes. Most economists are very keen that we should separate the interwar years, that is, those between the establishment of the Fed and the end of World War II, from other years. That is, they like to treat the interwar period, the first decades of the Fed, as if it was practice, right? And you can understand why, especially if the goal is to apologize for the Fed, because everyone admits the interwar period was a lousy period, it was the worst period in the performance of the U.S. economy. Let's hope it stays the worst period.

6:48That certainly is true with respect to unemployment. You can see the average rate of unemployment mainly thanks to the Great Depression is much higher than in any other period. But even if we focus on the post-war period, well, we find a slight improvement in the period between the end of World War II and the final abandonment of the gold standard or closing of the gold window in 1971. But since then the unemployment rate has actually averaged a much higher level and of course the recent developments are only going to make it worse. Now we've got changes in unionization, minimum wage laws, a million other factors happening apart from monetary policy, which is why I don't want to put too much emphasis on this. But at least a bare comparison of the statistical record shows that we can't say that the post-Federal Reserve era has somehow ushered in a period of much lower average unemployment than before.

7:46And indeed, again, that's true if you just look at post-World War II, if you throw in the interwar practice years, the Fed's promise has been even more glaringly broken. I do think we should pay a lot more attention to the question of price stability, the behavior of prices, because everyone understands that the behavior of the price level and the inflation rate, that is something that the Federal Reserve can be held responsible for almost exclusively. Here we have a nice quote from Bernanke telling us about the importance of stable prices and how they allow the dollar to serve as a measure of value in making long-term contracts, engaging in long-term planning, or borrowing and lending for long periods.

8:35So, given that this is very important, how good a job has the Fed done? Well, it's notorious to all of you, of course, and to most people, that with respect to preserving the purchasing power of money, the record of the Fed has been appalling. Here you see statistics for the CPI, and I didn't draw a vertical line here, but the creation of the Fed would be around there. What's interesting is, if you look at the CPI, between 1790 and the Fed's establishment, the difference in the price indices is about 8%. The price index is about 100 in 1790, then the index would be 108 or so in 1914.

9:21A very trivial difference. Now those statistics are very, very crude and rough, but they're as good as anyone has come up with. or alternative measures we could point to, but they all tell essentially the same story. Almost from the beginning of the Fed's establishment, and a later slide will make it clear, the value of the dollar has inexorably gone down, with the notable exceptions of severe deflations in the Great Depression or interwar period. Let's look a little more closely. This is what's happened to the value of the dollar since the Fed's establishment. So this is the purchasing power of the dollar rather than the price level so it obviously goes in the opposite direction.

10:06And one of the things that's worth noting is that the greatest decline in the dollar has happened or the greatest declines have tended to coincide with periods when the Federal Reserve had the restraints imposed on it by the gold standard relaxed. This next slide will show it more clearly still. Here you have monthly inflation. What many people are unaware of is that the worst inflation episode in the post-Civil War period in the United States was not the one in the 1970s and early 1980s. It was actually one that took place immediately after the Fed's establishment during World War I, where monthly rates, some of them, of quarterly rates of inflation, if you annualize them, for some quarters were as high as 40%, equivalent to a 40% annual rate of inflation, which is several times higher than the highest annualized inflation rates of the 70s. Now what's going on there, I'm referring to the inflation over here, what's going on there is that there's an embargo on gold exports. We didn't go off the gold

11:19The gold standard, but the restraints on Federal Reserve money creation were relaxed considerably compared to the rest of the period prior to the Great Depression. The same thing happens in the 70s. You get high inflation following the abandonment of the gold standard. Here you have also some inflation in connection with the Second World War. The basic story is the more you relax the constraints on the Fed imposed by the existence of some kind of gold standard, the more it takes advantage of it to contribute to the erosion of the purchasing power of money. Putting it the other way, to the extent that the Fed has in some parts of its existence resisted allowing the value of money to decline, it has done so mostly while it has been constrained by the gold standard.

12:14The gold standard research tank for what little preservation of the purchasing power of money has taken place since the Fed's establishment, not the Fed itself.

12:27Bernanke's statement emphasizes long-term contracting and the need for people to be able to predict what's happening to the price level. How good a job is the Fed done with regard to the predictability of prices? It's important to see that the high inflation or rapidly eroding purchasing power of money doesn't necessarily mean a more difficult to predict inflation rate or purchasing power of money, right? And some economists have argued that, well, even though the inflation rate has tended to get worse since the Fed's establishment, inflation and the price level have become more predictable, right? The inflation rate tends to be high but you know where it's going to be whereas before the Civil War it was harder to predict the rate of inflation.

13:17There's some truth to that but the truth to it is such as really doesn't mean that it was easier to make long-term contracts before the Fed's establishment, quite the contrary. Under the pre-fed gold standard, as we've seen, the price level in the long run didn't change much at all. Well, that implied that if you did have some increase in the price level, chances were you were going to have some downward movement to offset that in the future. And that's exactly what happened in the gold standard period. So you couldn't predict, it wasn't a good guess that the inflation rate tomorrow would be like the inflation rate yesterday. It might well be negative, the inflation rate yesterday. But in terms of predictability of the long run price level, that isn't the right question to ask.

14:03The right question to ask is how much uncertainty is likely to be connected with your guess of the price level five years from now, ten years from now, a hundred years from now, right? How accurate will our forecast be for those long horizons if we're trying to say make investments for those periods? What this chart shows, I won't go into the boring statistical details, but this shows, as it were, a measure of the price level uncertainty for different forecast periods or horizons. First under the pre-fed, pre-1914 period, and the data here start in the 1870s, and then we looked at the uncertainty for a five-year horizon, a 30-year planning horizon, and then a 100-year horizon.

14:54And over here, we look with the same scales at the levels of uncertainty during the post-World War II period. We're leaving the interwar period out again to give the Fed as much credibility as possible. And as you can see, the further you go out in time, the more, I don't know what's happened to my, oh, I've covered it. The more your predictability declines, what does this mean in practice? It meant that, whereas during the gold standard pre-fed period, especially in the last part of the 19th century, all kinds of companies, railroad companies and other companies were issuing 100-year bonds, right?

15:42Because there was enough predictability out there, there was enough certainty, of course There was enough certainty based on past experience that people felt confident that they knew what the price level was going to be in 100 years, confident enough to issue 100 year bonds. This picture, if you believe it, would lead you to conclude that the market for such long-term bonds would completely dry up. In fact, it did, completely dried up until the 1990s when a period called the Great Moderation of Relative Stability somewhat revived the market for such long-term securities. Nevertheless, the overall record of the Fed has been not to make it easier for people to invest or to keep interest rates low for long-term investments, but just the opposite. By the way, the current situation surely is one where what little revival in long-term lending markets had taken in place in the 90s. I'm sure that that is just about drying up again. If anybody's issuing 100-year bonds right now, well, I want to know because I might want to buy some. Now,

16:54some people will tell you, a lot of economists will tell you that, well, even though the Fed's allowed persistent inflation for most of its existence, at least it's prevented deflation most of the time. And that's good because deflation is really, really bad. And The gold standard, in contrast, and the pre-fed arrangement that went with the gold standard, that allowed a lot of deflation, and therefore, even though the feds allowed inflation, it has made up for that by preventing deflation, which is even worse. Well, to confront this argument, if the first of all recognize what a lot of economists don't recognize, but good ones, and not just Austrians, but good mainstream ones do, there There are two kinds of deflation, good and bad. Good deflation, well bad deflation is this kind. Here you've got the supply of goods, you've got the prices and output. If there's a decline in the money supply, for example, a sudden shrinkage, the demand schedule collapses, or a financial crisis like we had in 2008, you will get a fall in prices and that fall

18:01People and prices will be associated with reduced output and unemployment, the classic deflationary crisis. The Great Depression, the beginnings of the Depression in the early 30s fit this pattern. It's the kind of deflation Bernanke worries about, but he also tends to assume it's the only kind. Let's just accept that this is bad. And by the way, if you do, remember it's not the deflation that's bad. The deflation is the economy's way of dealing with a collapse of output. If there were more deflation, you'd actually be better

18:59The downward movement prices is associated with improvements in output, increased output. What about the Fed's record with respect to deflation? The deflation that took place in the pre-Fed period, with short exceptions, was mostly benign deflation. It was mostly driven by improvements in supply. So mostly good deflation. After the Fed's establishment, good deflation pretty much disappeared. There's hardly any episode where the Fed allowed prices to fall in response to improved output, even though output did improve for much of the Fed period. The Fed offset that with money creation. But that bad deflation actually became, if not more common, it became worse. That is, there were a few episodes of bad deflation that were the worst episodes ever under the Fed's term.

19:56The most notorious of the early 1930s, another one before that in 1920-21, recently a very bad episode of bad deflation in early 2008. So what the Fed's done is it wiped out the good kind of deflation, it gave us the worst ever episodes of bad deflation, and then whenever it wasn't doing that it gave us bad inflation. So overall the record on the price level, and again this is in terms that any mainstream that any economist should be able to accept has been very bad. Speaking of the mainstream, here's an article that appeared in the American Economic Review, a conclusion of it lately. It says, a broad historical look finds many more periods of deflation with reasonable growth than with depression, and many more periods of depression with inflation than with deflation.

20:44And what are the exceptions? Well, they're all after the Fed's establishment. Here is the bad deflation of the early 30s, here is the recent one, you can see the percent change in CPI and what's happening with output at the same time. All right. Let's talk about stability of output. One of the other prominent goals of the Fed and supposed accomplishments of the Fed is to stabilize the economy. Now we're just going to look at plain old statistics of the variability of output. Couldn't be cruder about this, but this is how most economists think about it.

21:31Look at what's happening in the standard deviation of output. Is it better before the Fed or after? Most economists, until recently, would have said, well, the Fed has improved things. Well, okay, not between the wars. That was practice. But since World War II, output has been more stable than before the Fed was created. And indeed, the early statistics, the statistics that everybody relied on until about 20 years ago, do suggest that. So here's, this shows you, I'm sorry, I didn't write those vertical lines. Here's the Fed's establishment. Here's that nasty interwar period. We leave that out, which is a big concession, isn't it? We leave that out, practice, and we look at the standard deviation here and compare it to over here. This is smoother.

22:20Except, those pre-war output statistics, turns out that they are deeply flawed. and we know that they are deeply flawed thanks to this person. That's Christina Romer and she was until last August the head of the President's Council of Economic Advisers. Romer did some close examination of the old statistics used for output in the United States in the days before 1920 actually and she realized that what people had done was they had used commodity prices because they didn't have prices for a lot of other kinds of output and then they assume that all prices kind of fluctuated as much as commodity prices. Well, she noted, if you look at commodity prices and other prices since the Fed's establishment, when we have good statistics on both, commodity prices are much more volatile than other prices, about five times as much. So these old statistics would have grossly exaggerated the volatility of the output measures, right?

23:32So she goes back and she comes up with her own statistics modified for the difference between how much commodity output fluctuates and commodity prices compared to other kinds. And this is what her standard deviation measures look like. And they make a huge difference, because it turns out now, if you look at the average variability of output before the Fed and after, it's about the same. And in fact, depending on how you take the trend out, because you have to take the trend out, right, to come up with a standard deviation, if you do it in a somewhat, in what most economists consider a more sophisticated way that allows for a varying trend instead of just a linear one, a terministic trend, the pre-war period looks more stable, the pre-Fed period looks more stable from the post-World War II period. More stable than the post-World War II period. Throw the interwar period in and the Fed's overall record in terms of sheer stability of output is much

24:35worse than the pre-fed period. Now, notice the pre-fed system stank. It was a lousy system. So when I'm saying these things, I'm not saying to you, oh, we should go back to the pre-fed System. What I am saying is, the Fed cannot claim that it has accomplished, even since World War II, more stability than the old system. Now, you would be right if you said to me, and a lot of economists would make this point, of course, say, well, look, you know, a lot of things have changed since 1914, besides the fact that we now have a Fed, but You can't just look at these statistics you've been looking at, for example, about the fluctuations in output and say, well, the differences are only because before you had no FED and now you do. Absolutely. There are a lot of other changes that have taken place in the structure of the economy. However, looking at most of these changes, what you find is, well, looking

25:37at as many as I can think of, what you find is most of the changes either shouldn't have There is one major structural change that I can think of, that certainly would have made a difference, apart from what the Monetary Regulatory Commission has done in the last couple of years. It has made a difference in the kinds of output fluctuations that we've just been looking at. That is, there's no reason to think their effects would be neutral. Example, we have a lot less agriculture. We've got a lot more services. Manufacturing actually hasn't changed that much. So those shares have changed, but there's no reason to think that that shift would have made much difference in overall output stability. from what the monetary regime was. And that is the size of government. Mainstream economists will tell you that because government hands out welfare payments and other things that are either acyclical, those payments don't decline when output generally declines, or they're counter-cyclical, they actually go up when output declines,

26:32that the bigger government gets relative to the whole economy, the more stability you should have in total measured GDP, right? They don't necessarily say, therefore you should have big government, some of them do, because it also is true that you get less total output as government grows, according to the statistical studies that show the stabilizing effect of more government. In any event, government's grown tremendously since the Fed was established. It's not counting the recent burst of government growth. It's about five, there's about five times as much government spending as a share of GDP today, or say in early 2008 than there was in 1914, which according to some studies should have meant a tremendous improvement in macroeconomic stability, that is, stability of output.

27:19So if you controlled for that, you should have expected the Fed, we should expect the post-Fed era, even if the Fed had done nothing to change stability, to show a lot less instability and the pre-fed era. Yet it is about the same if not slightly worse in the post-fed period. So something is offsetting the tendency of big government to stabilize GDP. And it's obviously, one possible candidate is the Fed. The Fed's actually been destabilizing a lot. Another thing that we can look at is the frequency and length of business recessions or contractions.

28:05Now, for better or worse, the official body that has always decided when these are taking place and how long they last is the National Bureau of Economic Research. And of course, they're the ones who assure us that the most recent recession ended June of 2009. According to the same people, the frequency and length of recessions before 1914, there were many more recessions and they lasted a lot longer before 1914 than since World War II. We leave out the interwar period because that was practice. Right. So, now, the way in which the NBER dates cycles is quite mysterious, and I think it might be mysterious even for the people in the NBER.

28:58And it's mysterious the way they do it for recent cycles. The further back you go in history, though, the more bizarre their dating methods become, The point where, once again, Christina Romer, who's done a lot of research on this, had went back and determined that actually their dating of pre-fed business cycles was completely bogus. So she did some re-dating and she came up with this conclusion, based on her own more careful statistics. And this is Christina Romer. This is not, she is not a Mises fellow. This is in 1999, this quote. That's very significant, right? Recessions have not become noticeably shorter over time.

29:46The average length of recessions is actually one month longer in the post-World War II era than in the pre-World War I era. There's also no obvious change in distribution of the length of recessions between the pre-war and post-war eras. Christina Romer, 1999. Why am I stressing that? Obviously because if you add recent experience, that conclusion is going to get stronger, isn't it?

30:16Just to give you an idea of how much difference it makes when you recalculate these statistics, another economist went a little further back at Romer and looked at, among other things, at the crisis of 1873. According to the NBER, that thing lasted for six years. It was a recession that lasted from 1873 to 1879. The longest in history? Well, according to the revised statistics, it lasted two years. A little longer than the official duration of the most recent recession. So, obviously, a lot depends on which statistics you're using. One more factor worth considering. When we talk about the fluctuations in the economy, we should really consider what the sources of those fluctuations are.

31:12Economies can be exposed to either shocks to supply or shocks to demand. Now, we expect the monetary system to minimize the latter. Shocks to demand, shocks to spending. A well-functioning monetary system will tend to stabilize those. If the demand for money goes up, the supply can go up somewhat and keep spending on an even keel. This is, by the way, the standard view. If you have a Misesian alternative view, that's fine, but we're making an imminent criticism. On the other hand, supply shocks can happen that Monetary authorities are powerless to do anything about. For example, bad harvests or a war. Well, we should therefore try to take account of the different incidents of supply and demand shocks in the pre and post-Fed period to get a true grasp of how much the Fed has contributed or detracted from stability.

32:06Well, if you do, statistically, you come up with this. In the pre-Fed period, supply shocks were overwhelmingly responsible for fluctuations in output. They explain something like 90% of those fluctuations. In the post-fed period, depending on which horizon you look at, they explain a much smaller percent. In other words, the pre-fed period was one where the monetary system could not be expected to produce much more stable output because there were big supply innovations and harvest failures and other things that were quite independent of the performance of the monetary system that were causing output to bounce around back then. Since the war, those fluctuations have been relatively much less important.

32:54So once again, the bare statistics exaggerate the success of the Fed. And if we have to adjust, we need to adjust downward our assessment of the Fed's performance, its contribution to economic stability.

33:10Banking panics. Well, a lot of people feel that the Fed as a central bank capable of acting as a lender of last resort has helped at least to minimize banking panics. You don't need to be told in this room that the very worst banking panic of all, that of the early 1930s, took place during the Fed era. It was much worse than the one in 1907 that prompted the creation of the Fed in the first Here, though, is an overall assessment of the record of the Fed with respect to the pre-Fed era with regard to banking panics. Elmer Swicker knows more about this than anyone else. And as you can see, he concludes that there were no more than three major banking panics between 1873 and 1907, two incipient panics in 1884 and 1890, twelve major panics between 1873 and 1907, Panics in 1884 and 1890, 12 years elapsed between the Panic of 1861 and that of 1873 and 20 years between the Panics of 1873 and 1893 and 14 between the Panics of 1893 and 1907.

34:21Three banking panics in half a century and only one of three, 1893, did the number of bank suspensions match those of the Great Depression. During the early thirties, Wicker also writes elsewhere, there were five major panics in a row. Now, panics did stop, but they stopped after the holiday of March 1933. This figure shows the total number of bank suspensions. As you can see, they actually go up in the early years of the Fed, and then they stop almost altogether. The bank suspensions here are a rough measure of bank panics. When you have enough of them, it's a panic. How's that? They do stop after 1933, but what's behind that?

35:06Not the Fed. When the Fed existed but the FDIC didn't, bank failures are getting worse and worse. It's the FDIC, deposit insurance, that for a while is putting a lid on major bank suspensions. and we know why it did that it allowed the banks essentially to be guaranteed at the risk of the exposing taxpayers ultimately of course that system would itself come unglued so I'm not recommending FDIC I'm pointing out to you though that the statistics clearly show that it wasn't until the advent of the FDIC that bank panics and suspensions at least for a time went away, the Fed itself did nothing to prevent the incidence of panics.

35:56Here you can see that the good news didn't, that the success of the FDIC doesn't go on into the recent era. What about the recent crisis itself? Well, Bill Buder, who's a very good economist, again not a Mises Institute type at all, concludes is that the interventions during this crisis, far from having contributed overall to the stability of the U.S. economy, appear to have been designed to maximize bad incentives for future reckless lending and borrowing by the institutions affected by them. Indeed, I think that the Fed, probably the worst things it's done in its whole history are the bailouts it's engaged in in the recent crisis because they are setting the stage for an even greater moral hazard problem in the future and even more disastrous crises in the future. Central bankers will like to remind you or like to claim that the great English journalist and economist Walter Badgett was the one who said that we should be grateful to have them as lenders of last resort. Badgett wrote a book called Lombard Street where he

37:10He did recommend that the Bank of England act as a lender of last resort, and that's where the idea for having central banks do the same elsewhere came from. But what most economists don't tell you, and most of them don't know, is that Badgett in writing that book admitted that it would be better not to have a central bank at all, that it would have been better if England never gave monopoly privileges to the Bank of England. He recommended and the last resort lending rule as a way to minimize the harm done by the Bank of England to a system that it had rendered unstable. And here's his concluding passages from Lombard Street where he makes it very clear that although I've suggested, pointed out a deep malady and only suggested a superficial remedy, the lender of last resort, I've tediously insisted that the natural system of banking is that where many banks keep their own cash reserves, and so on.

38:07So, Badgett, I'm going to rush through this, but Badgett did make this recommendation, but it was because he thought at that time you could never get rid of the Bank of England, so it was the best solution he could come up with. Do I want to push back the clock? Is that what I'm arguing for? I'm not. I'm not saying that the old system before the Fed was ideal or that we should be happy What I am suggesting though is if people thought in 1907, after the panic of 1907, that the system isn't working and we need something else, they have just as much reason, if not more, for thinking that today. Our system has not gotten better since the panic of 1907. The Fed has not delivered on its promises.

38:56I was going to say something about Canada but I don't have time and about the gold standard. But I will say something about the Wizard of Oz. What I really hope to accomplish with this kind of talk, and with a long paper I've written with some colleagues that's aimed at the mainstream economists, isn't though to propose any particular reform, but to get people away from the complacency about the Fed and thinking of it as the only system that we could possibly have and one that is at at least better than anything we've had before. Whenever I think of Ben Bernanke or Greenspan, I always think about this scene in The Wizard of Oz, where there are wizards there and you've got the steam and the mirrors and the colored lights and all that.

39:46And I see myself as sort of the Toto of Federal Reserve critics. I want to move that curtain away and expose this institution, revealing the fact that it is a very frail, Very unsuccessful arrangement. Not something we should regard as the last word in monetary regimes. So with that, I will close my talk and thank you all once again. Thank you very much.