 Welcome to the second lecture this morning. Our topic is central banking and inflation. And I will take the next 45 minutes to explain to you why we need central banks to protect us from inflation. So of course, that's not my thesis, quite the contrary. Here's the plan for the lecture. I will, as usual, introduce into the topic. I take a couple of minutes to define some basic notions. And then I want to talk about measures and measurements again, and here specifically about some problems when it comes to inflation measurement. This is the most technical part of the talk. I go then into the consequences of inflation, in particular, the redistributive effects of inflation. I will argue that monetary policy, monetary expansion, as conducted in recent decades in most countries, leads to more inequality and undermines social mobility. And I will then conclude with some further implications of this development. Again, if you want to read about the topic, I have another recent publication out with a co-author from Germany. This paper is a case study specifically on Germany. You will find some elements of the presentation in that paper in more detail. I have added to this presentation some other empirical examples from France, which is also a very spectacular case, and some empirical developments from the US as well, since we are in the US. All right, let me start. So central banks, what are central banks? We have all a intuitive understanding. They are the institutions in charge of monetary policy. In the modern fiat money system, we can say that central banks are essentially defined by a legal monopoly in the production of base money. And by base money, which is often also called M0, we mean cash in circulation, so cash in private hands, for example, in your wallets or under your pillow at home. And we mean bank reserves. So that's cash stored at the bank or deposits that the commercial banks have with the central bank. This is M0, and this is under direct control of the central bank. And as I said, central banks are in the business of monetary policy. And when we talk about monetary policy, it is always about changing M0 to influence the somewhat broader monetary aggregate, which is M1. M1 is only indirectly controlled by the central banks through monetary policies. And among monetary policies, we have, for example, legal reserve requirements. Not very important anymore. We have interest rate policies. Used to be very important. Then when interest rates were at zero, nobody talked about this anymore because there was no way really to go further. To some extent, we had negative interest rates. But yeah, they have become more important again with the rise of inflation and the increase in interest rates in order to counter inflation. We have open market operations to control for the amount of reserves that banks hold. And we have unconventional purchase programs, which are basically open market operations on steroids, where central banks buy massive amounts of assets and thus increase the amount of reserves and the amount of M0 in the economy. So what is the goal of monetary policy? It depends a little bit on where you are. But generally, the goal is an inflation target. So in Europe, in the eurozone, there is a primary policy goal of a 2% inflation rate year on year, measured by the HICP. And then now all other monetary policy goals are secondary to that primary policy goal. In the US, at least on paper, it is somewhat different. And the inflation target is on the same level with other monetary policy goals, such as employment, real economic growth, and things like that. But it's important to note that the goal of monetary policy really is not price stability, but price inflation. And 2% price inflation has been deemed to be optimal for some reason. I never understood that. In fact, I remember vividly a conference that I attended in Germany, the Nobel Memorial Conference in Lindau. This is an occasion where Nobel laureates come and talk to younger researchers and students. And there was a panel with Christopher Sims. And Christopher Sims was asked, what are the problems in Europe and what should we do? So that was the time a couple of years after the Great Recession. We had the immigration crisis. We had the European debt crisis. All of these problems, very troubled times. And Christopher Sims said, well, price inflation is too low. It has to get back to 2%. And most of the problems would be solved. Highly doubtful. But the target anyways of monetary policy is this 2% inflation rate. And of course, that raises the question of how to measure inflation. That's the big problem, really. Inflation in the classical sense of the word is just an increase in the money stock. And that would be a rather simple and, in fact, objectively measurable type of inflation. Now, we can, in principle, measure exactly by how much the money stock is expanded once we have a definition of the money stock every year. But of course, inflation in the modern sense of the word means an average increase in money prices. That's why we sometimes call it price inflation instead of only inflation. And that immediately raises the question of what prices should we look at, all or not all, just a selection of prices. And then whatever it should we calculate, how should we calculate it, how should we weight these prices. And yeah, the ultimate goal of this price index that we come up with is to assess or estimate how much more money one needs in order to obtain a given level of one satisfaction or a given standard of living, really. And you see that this problem is ultimately unsolvable, which should be clear from my first presentation, right? It's, again, about welfare and utility, something that is not measurable. So we have to come up with some best approximation. But really, it is very difficult to do that. And there is really no objective level of prices, no objective price level, and no objective rate of inflation. This is not only because individuals are different and have subjective evaluations of things. Even for one individual, you couldn't calculate objectively a price level and an inflation rate. Because it's not clear how you would weight the different goods, even for one individual. It remains always, to some extent, arbitrary. So that has not prevented monetary policymakers from declaring price inflation to be the primary goal. And here you can see the performance, roughly. Since the introduction of the euro, you have here plotted the ECB target rate of 2%. And you see the inflation rates measured by the HICP in Germany and France. And they have fluctuated more or less around these 2% and have increased recently in 2022. On average, the 2% have been reached. So that's the great news, right? With the recent HICP, we are actually now quite good on average. We have 2%. Before that, it was, on average, below 2%. And that was the big problem that Christopher Sims saw. And that was, of course, also the justification for why the ECB and the European Central Banks increased the money stock in the eurozone quite substantially. Here you have the same sort of picture for the US. It's very similar. It's a bit more, it fluctuates a bit more in the US. There's a bit more going on in America. It's a wilder economy, probably. And on average, inflation is a bit higher. It's a bit above 2%. It has also recently hiked. You all know that. Now let us look at the money stock M1 that I already introduced and how it has evolved in the eurozone. As I said, the relatively low price inflation rates below the 2% target on average before the recent HICP have been a justification for expanding the money stock by massive amounts. Here you can see, plotted as an index, M1. And you see that over the course of time, since the introduction of the euro, that money stock has increased by a factor of larger than 6%. It's quite substantial, quite substantial increase. And a good benchmark for comparison is to look at how the real economy has developed at the same time. And when you add the real economy, you can see that the performance is not at all impressive. The real economy in the eurozone has grown by something like 40%, if at all, over the entire period. So a massive increase in the money stock, only a very moderate real growth of the economy. And so you end up, if you do the calculations, with an average excess money growth above real economic growth of 6.8 percentage points every year. So 6.8 percentage points faster money growth than real economic growth. That would also be somewhat of a measure of inflation. And you would think that if the official inflation measure that the ECB is using is somewhat accurate or is reflecting general price inflationary developments, then it should be closer to 6.8% than to 2% on average. So this is some evidence, suggestive evidence that the HICP is maybe not quite reflective of general price inflation. We can do the same for the US. But here we have something very special. Again, you think that now maybe that in the course of COVID, the Fed has gone completely crazy. But this increase that you see on the right-hand side of the graph is also due to a change in the definition of M1. So I don't think that was a strategic choice. But they now can say that this increase is because of a change in definition, not because of expansionary monetary policy, or at least not entirely. If we cut that out and if we bring that plot to roughly the same time frame as before for the eurozone, the plot looks like this. So we have a similar development. Monetary growth is much higher than real economic growth. And if we do the calculations once more, we get an average excess money growth above real economic growth of 5.6% per year. So it's also quite substantial. And there is evidence that the CPI as calculated here in the US is not quite reflective of general price inflation. Now, we want to think about what sort of explanations are there for this gap between money growth and real economic growth. And there are essentially two explanations. Of course, it is possible that something happened to the demand for money, right? Maybe at the same time the demand for money has increased. That is the reservation demand for money. People just hold on to the money and don't spend it. And if they don't spend it, then, of course, there is not price inflation to that extent. But the second possibility is that the price index, the CPI in the US or the HICP in Europe, underestimating price inflation. Now, why could this be? Why does the index underestimate general price inflation? There are, again, two possibilities. First of all, the index could be downward biased. It could be the case that the statisticians do their very best to estimate inflation, but they just don't get it quite right. The index, as it is calculated, is giving us a number that is just too low. It could also be the case that the statisticians don't make any noteworthy structural mistakes. They are good at what they try to measure, namely consumer prices. But there is inflation outside of the basket of goods and services that is used for these indices. And I want to talk a little bit about both possibilities. I think that the first general explanation of the gap could play a role as well, but it certainly does not explain the entire gap. So there is some room for thinking about why these indices do not quite reflect price inflation correctly. And I want to go over both explanations for why that is. First of all, why could the index be biased? And when you go into the literature, you find two main sources for bias in price indices. Those biases are due to substitution effects. That is, people change their consumption behavior. They change the basket of goods and services that they consume over time. And then there are quality changes to the goods and services. This is also considered to be one of the main causes for bias in inflation measurement. So let's first talk about substitution effects. It was in the 1990s in the US that the Boskin Commission estimated that because of substitution effects, the official CPI is about 0.4 percentage points too high. It's overestimating inflation. We have to correct our measurement techniques, our calculation methods, in order to correct for this. What the Boskin Commission observed was that consumers switched from products with relatively high inflation rates to products with relatively low inflation rate. Why is that the case? One wonders. And because they switched to these products with lower inflation rate, we should increase their weight in the overall index and decrease the weight of those other products with higher price inflation rates, and thus we would have a lower average price inflation rate. So to illustrate the point, I give you an example. Think about two goods that are considered to be substitutes like Coca-Cola and Pepsi-Cola over a given period of time. Think about a situation where Coca-Cola has a price increase of 1% over that period of time. And Pepsi-Cola has an increase of 5%. Now, when we want to weight these different price developments, we look at the expenditure share of consumers on those products. And now think about a situation where the initial expenditure shares are 50-50. People spend about as much money on Coca-Cola as on Pepsi-Cola. And then the updated expenditure shares after the adjustment in the consumption behavior is 80-20. Now they spend 80% on Coca-Cola overall because they now focus on Coca-Cola because price inflation was low. And they spend only 20% of their money on Pepsi-Cola because price inflation was high. Now the question is, what weighting scheme should we use? Should we use the old expenditure shares or the new expenditure shares? And if you use the old, the initial shares, then you get an average inflation rate of 3%. It's just equally weighted, 1%-5% on average, 3%. However, if you use the updated expenditure shares, you weight the 1% inflation for Coca-Cola with 80%, 0.8%. And you weight the 5% for Pepsi-Cola with 0.2%. You get an average inflation rate of 1.8%, which is much lower, and you are miraculously close to the target as well. Now the question is, what should you use? What inflation rate should you use? What is the correct price inflation rate? And there is no objective answer to this. Both of them are equally plausible. Irving Fischer had a genius answer to how to solve it. Just take the average of the two. And Gottfried Haberler, an Austrian economist, he analyzed this situation in a bit more detail. He argued that, well, yeah, you can't really know. But under certain assumptions, these two rates, which correspond to the partial and the Lasperes index, give you the upper and lower bound. So the truth may indeed lie somewhere in between. So Haberler gave some credit to Irving Fischer's proposal. Now the second source for bias are quality changes. And here the basic idea is that quality improvements on the product are like an implicit price decrease. Quality deteriorations are like an implicit price increase. Because you get, for the same amount of money, let's say, a better quality when there is a quality improvement, so this is like making the product cheaper, in a sense. Or you get a lesser quality, which is like making the product more expensive for you. So and this should be corrected. The Boskin Commission, again, analyzed this cause for bias and estimated that the CPI in the US overestimates price inflation because of quality changes by, again, about a half a percentage point. And so this was an argument to introduce more and more explicit quality adjustments. So instead of just taking the observed prices that you see in the stores, you correct the prices in cases of quality changes. The Boskin Commission said overall quality improves. And we do not adequately adjust for this. We should do that. And this would make price inflation lower. Again, I give you a little example. Think about a computer model that is sold for $1,000 and has a CPU of 3 gigahertz. And then there's the new model that replaces the old one. It costs 100 euros more. On paper, that would be a 10% inflation rate. But the CPU is faster, 3.3 gigahertz. So now what would the statisticians do? They would try to estimate the monetary value of a 0.3 gigahertz increase in the CPU performance. And they would do this with a regression over comparable products. And maybe they estimate that this improvement in quality is worth, on average, $80. And then they would deduct that from the new price. And the adjusted inflation rate is 2% instead of 10%. So why is that problematic? Well, that's problematic for many reasons. Of course, quality is subjective that we all know. But there's another technical reason why there is a likely bias towards taking account of quality improvements and not so much of quality deteriorations. Quality improvements are, of course, something that the producer of the product would advertise. It's very open, it's very transparent, that's very visible. Quality deteriorations are hidden from the consumer, are hidden from the observers. So it's much harder to take account of those. And they are typically done on other margins of the product. So it's not the CPU that is lower all of a sudden, but it might be something else. Maybe some material of lesser quality is used more plastic than metal or the duration of serviceability of the computer might be lower. Maybe it only works for two years instead of four, whatever it may be. It's harder to take into account. And that's why there's a likely bias in the statistical procedures to take account of improvements, thus lowering the price inflation rate and not taking account of deteriorations, which would lead to an increase in the price inflation rate. And my colleague and I, we had a meeting with the statisticians in Germany who are in charge of calculating the index or some representatives of the statistical offices. And we asked them, so could you give us examples of quality improvements that you take into account? Of course, and they gave us all these examples, computers, smartphones, everything. And then we asked, could you give us also an example of where you take into account the quality deterioration? And I kid you not, there was dead silence. They couldn't give us any example. Apparently, this is not done. Well, of course, that is a reason to believe that these adjustments that were made after the work of the Boskin Commission lead to a downward bias in the HICP. But it's very difficult to come to a definitive conclusion here because quality is subjective. What is not subjective, however, is the observation that the CPI, the HICP, focuses on only one part of goods and services in the economy. These indices focus on present private consumption. And accordingly, they leave other important things out. So for example, future consumption or saving goods that we would buy in order to save for the future, save for future consumption, for example, stocks in real estate are not included in the index. Also, public goods that is government spending is not included in the index. And it's very easy to see that there is over-proportionate price inflation in both of these areas. We have, for example, we have done the calculations for Germany for public goods. What is the price for public goods? Well, it's the amount of taxes you pay. The tax revenue of the German government has increased in the period since the introduction of the euro by much more than the general price inflation rate measured by the HICP. So there was an over-proportionate increase in the tax revenue of the German government. This could be considered to be an over-proportionate inflation for public goods. And here in this presentation, I want to focus on assets. And there, too, we see over-proportionate price inflation. And again, this is very easy to explain. Why do we have over-proportionate asset price inflation? Well, it is because inflation changes not only our consumption behavior, but also our saving behavior. We try to protect ourselves from inflation. So there is naturally a decrease in the demand for assets that lose purchasing power, such as cash and deposits. And you redirect your savings into such asset classes that are promising in protecting you from inflation. So for example, you buy stocks and real estate. And if you look into the official data in Germany for the time period since the introduction of the euro, we look at the standard stock price index. It has grown, on average, every year by 4.8%. Of course, that's very volatile. Sometimes it goes down for a while. But on average, over longer periods of time, there is in almost all countries over-proportionate stock price inflation. You might argue, oh, this is because there was good economic performance in Germany. Maybe. But that's not really the case. Economic growth was quite low compared to other countries in Germany. And you can see that it is mostly driven by inflation, by looking at specific sub-periods. For example, the COVID pandemic and the lockdowns. During that time, we had, of course, an initial decrease in the stock prices. Then we had a massive monetary policy response. And stock prices increased over that period, on average, by an annual rate of 8%. So that's not because of real economic improvements or productivity gains or expected gains in the future for German businesses. That's mostly because of inflation. And you could look also at other assets, this time real estate or housing. Here you have the official housing indices for France and Germany. Very interestingly, it's the difference between the two. In France, we have this massive increase after the introduction of the euro. Within eight years, house prices have doubled, quite substantial. And they haven't decreased again. They stagnated for a while after the crisis. And now, since recent years, they're increasing further. In Germany, house prices were stable before the crisis. There was no housing boom in Germany. But since the crisis, house prices accelerate. They have almost doubled since then. So how does this compare to consumer price inflation? Here you have the corresponding HICP indices for the two countries. And you can see that house price inflation was indeed above average. We have over-proportionate house price inflation. So that's the main takeaway here. It's quite normal in an inflationary environment that you have over-proportionate asset price inflation. And with that come a lot of implications for the distribution of income and wealth. There are distributive effects of this development, of this heterogeneous inflationary process that tend to lead to greater inequalities. And I would argue that there are four important redistributional channels, so to speak. So the first one, of course, and over the history, this is probably one of the most important channels of redistribution. It is from the private sector to the public sector. That is why we have central banks, why we have monetary expansion. It facilitates government finance. So there is a redistribution from private to public. Inflation is like an additional tax that the citizens pay. Now for the question of inequality within society, the other channels are somewhat more important. There is, of course, a redistribution implied in over-proportionate asset price inflation from the poor who don't own assets to the rich who do. That's quite natural. If you own assets already, you enjoy a positive wealth effect from over-proportionate asset price inflation. If you don't own assets, it is harder for you to catch up. The distance to those who own assets measured in money is longer now. So the inequality increases. There is also a redistributional channel, I would argue, from labor income to capital income, and most importantly, capital gains. That's again related to asset price inflation. This is somewhat more controversial, especially among libertarians. On the left, or friends on the left, they really like this. They would totally agree with it. But when you talk to libertarians, you have to explain that some more. I would argue that labor income in particular is relatively reduced in the inflationary process, because an inflationary environment leads to a crowding out effect in the investment behavior. General price inflation and, in particular, over-proportionate asset price inflation makes speculative investments more attractive. That's somewhat of a misnomer, because all investments are speculative. We know that. But what I mean by that is just investments that are targeted towards price increases of the assets that are bought. It's not really a productive investment where you produce real capital goods that increase the capital stock of the economy. And it is those latter productive investments that ultimately are necessary to increase wages and labor income, because it is real capital that makes labor more productive and so on. And when you have this crowding out effect in an inflationary environment, labor income is harmed. And of course, you have benefits to capital gains because of over-proportionate asset price inflation. And then lastly, there is a redistribution channel that is intergenerational. There tends to be a redistribution from the young generations to the older generations. And that's also easy to understand once you accept the initial distribution channels, redistribution channels. Of course, it's young people, the young generation, that is particularly dependent on labor income and does not yet own assets very often. The older, on the other hand, have assets. They receive additional capital income and they benefit from a positive wealth effect in the face of over-proportionate asset price inflation. Lastly, and when it comes to social mobility, over-proportionate asset price inflation has an important effect because it increases, it drives up the wealth-to-income ratio in society. And the wealth-to-income ratio is a standard measure for social mobility. How difficult is it to climb up the social ladder? This can be estimated by the wealth-to-income ratio. And over-proportionate asset price inflation increases that ratio and thus undermines social mobility. So let us look at, again, our two examples here, France and Germany. You can see that there's a positive trend and it looks suspiciously similar to what we've seen in the house price indices. And this shouldn't surprise anyone. The implications of this development, when we look at France, for example, we have a wealth-to-income ratio of about 3.5 in the 90s. And then with the introduction of the euro, this wealth-to-income ratio increases to six, meaning that overall wealth of the French population is six times as large as annual income of the population. What does that mean? You can make a back-of-the-envelope calculation with a wealth-to-income ratio of 3.5. It would take you 35 years with an average income and a saving rate of 10% to obtain the average level of wealth in society if you start from zero. 35 years, average income, 10% saving rate. When the wealth-to-income ratio is at six, it would take you 60 years. That's much more difficult within one working life. And in recent years, the wealth-to-income ratio has increased even further to seven. It is now roughly at the same level as it was before World War I. Now there's a debate about why what is the cause of this development. Thomas Piketty has a very important paper out on this issue. And he argues, well, this positive trend in wealth-to-income ratios in the second half of the 20th century is mostly a recovery process from the world wars. But when you look at France, you see that it only really increased with the introduction of the euro. And it perfectly coincides with house price inflation. So you wonder, has France only recovered from World War II with the introduction of the euro? Probably not. Here you have the same data. And I interposed the house price indices just to illustrate the close correlation. And of course, correlation means causation. Here, just to see more clearly the dates, the introduction of the euro and the great recession. I have picked France for obvious reasons, I live there. But you could take other Southern European countries in particular, Spain, Italy, Portugal. They've all made similar developments. And Germany is sort of an outlier in this development. In Germany, house price inflation has accelerated only after the crisis in recent year with the advent of unconventional monetary policies. But you can see also here that in Germany, with the increases in house price inflation, the wealth-to-income ratio is pushed upward. And this also undermines social mobility. In Germany, of course. Here you have the data for the US. The case is somewhat less spectacular, but there's also a positive trend in the wealth-to-income ratio in the US. The big advantage that the US still has is that they allow incomes to be relatively high, which is why the wealth-to-income ratio is relatively low. It's lower than in France. It's much lower than in Spain. Because incomes are not taxed as heavily in the US. But you can also see here the positive correlation, of course, with the housing bubble. This pushed up the wealth-to-income ratio. Then you had a drop in the wealth-to-income ratio. Now recently, again, a hike. So the same developments, maybe to a slightly lesser extent in the US. Now, what are the further implications of this development? If you accept the arguments that monetary expansion, monetary policy as conducted in recent decades contributes to inequality and undermines social mobility, then there are a lot of further implications for society and the economy. And one implication I would argue is that there is a sense of injustice that spreads in society. When people see increasing, or just perceive, they don't really see it. They feel like there is more inequality. And maybe they even somewhat intuitively feel that this inequality is unjust, because it is an inequality that arises out of an inflationary process, then the sense of injustice that spreads is, to some extent, justified. If inequality emerges out of productive economic activity, OK, most people intuitively, a common sense of justice wouldn't be against that. But when it is actually a process that is inherently unjust, when it is the inflationary process that leads to more inequality, then, of course, the sense of injustice is justified in a way. And this then contributes to some degree of resentment towards public institutions, to the politics, to the economic system as a whole. And again, that seems to be justified when this process that leads to this inequality is inherently unjust. Well, what could be done about it? Of course, something would have to change with the system, but the problem is that the people who benefit from the redistribution process are also the people who are capable of changing the system. But it's not in their self-interest to do so, because they are benefiting from it. And that leads to some degree of collective corruption as Thorsten Pollard called it in a publication in the QJAE from 2011, I think. I think that's a great term. There is a degree of collective corruption, even if the problems are perceived, if they are known, they are not admitted in the public discourse, because it's not in the interest of those who can change it. So this has further implications for politics, for example, you have decreasing voter participation. If you see the system being rigged against you, and if you have the feeling that the elites are not really on your side, if you have the feeling that there is some degree of collective corruption, then you are not interested in participating in the democratic process and voting. So you have decreasing voter participation, and you have a shift towards politically more extreme parties on the left and the right. To some extent, you might argue that, well, it's great, right? People vote less, they are less interested in politics, that's OK, that's great. But I believe that there is a certain danger in this process as well. If you think about tyrannical systems in history, they all emerged out of a situation where there was deep discontent with the existing institutions. And it's quite striking. Official survey data from the US shows that in the 60s, more than 70% of people would say that the government almost always does the right thing. More than 70% in the 60s. Today, there are still 20% who say that. You might wonder, who are these 20%, right? But there's still a striking development. There is a loss in trust in public institutions, and that's quite potentially dangerous. There's also, of course, potentially a chance, for a change in a good way. There are other implications to this development. If, indeed, social mobility is undermined in the inflationary process, then, of course, it is sort of understandable that people might decide not even to try. If building up wealth has become so difficult, the opportunity costs of building up wealth for the future have become higher, why not just enjoy the little things you have right now in the present and consume what you have? So I believe that this inflationary process is contributing to more widespread, hedonistic lifestyle choices, more consumerism, less rugality. So those are things that, again, our friends on the left would criticize, would blame on capitalism. I think it's mostly because of monetary interventionism. Lastly, and you might think, well, all the bad things in life are due to monetary policy. Of course, I do not believe that these more broadly social societal problems are entirely due to monetary policy. There are many different causes of this development. But monetary policy and inflation might be a contributing factor to these developments. And also, I believe, to the more widespread existential fears that you perceive in society, people feeling like they can't make it, there is a despair among people, and particularly among young generations. If you look at the typical indicators for this psychological problems, suicide rates, and all of that, they are up, probably for many reasons. But the increase in inequality, the undermining of social mobility might be contributing factors in this development. So with these optimistic reflections, I want to close my lecture. And thank you for your attention.