## Wednesday, March 30, 2016

### Quantitative easing is better than helicopter drops

It's straightforward to induce inflation, harder to preserve the Lumo principles of safe and helpful interventions

Capitalist and Imperialist Pig quotes a left-wing would-be economist named Brad DeLong. This Gentleman superficially names the "global savings glut" and "deflation threat" as two main enemies and proposes to literally revive Milton Friedman's concept of the helicopter money. (They sometimes have the chutzpah to talk about the "austerity madness", too. But in this world in which most governments still run several-percent budget deficits, I would find it a self-humiliation to try to respond to insanities such as the "austerity madness" so let me ignore it.)

According to the plan, Janet Yellen will fill green helicopters with lots of money and pour them over the United States, especially above the neighborhoods with lots of colorful gay female poor uneducated discriminated Americans. This will increase the aggregate demand, the inflation rate, and everyone will be happy again.

The idea of "throwing money off the helicopters" was coined by Milton Friedman in his famous 1969 essays, The Optimum Quantity of Money. The monetarist economist whom I admire has argued that if the economic output is reduced by deflationary expectations (people delay consumption because the prices are expected to go down), it's straightforward to reignite or increase inflation. Just print the banknotes and throw them to random people. If there's a monetary problem (low or wrong money supply), it should be fixed and it may be fixed by monetary tools (tools that only care about the overall macroeconomic sums – the total money supply etc.).

Friedman has argued that if such an exercise takes place once – so that it doesn't change the expectations about the future – the exercise is a net benefit pretty much for everyone. Those who collect the banknotes from the ground objectively benefit but those who don't collect any banknotes don't get any poorer. Well, except for the fact that the unlucky ones may also care about their "relative wealth" and because the "banknote collectors" have gotten wealthier so easily, others may consider it unfair.

At any rate, I agree that such a one-time event is basically beneficial for the society. The real problem is that once something like that may occur, it may occur repeatedly and people may expect that such an event will take place again. And this expectation may transform them into people who are waiting for the money to fall from the heaven, thus reducing their useful economic activity. Increased helicopter expectations will also increase the inflation expectations and also the inflation itself, which eventually neutralizes all the good effects of the helicopter drops.

The helicopter drops don't have to be performed literally. If you reduce the taxes or increase some welfare programs and if you run public budget deficits that the country shouldn't really afford, and you start to "automatically" pay for this new debt by newly printed money, it's obvious that you're basically throwing the money out of helicopters and the random citizens will get them. The sickness of this exercise is obvious: the higher percentage of the money is obtained from the helicopters (literally or otherwise), the smaller is the correlation between useful work and one's wealth, so the motivation driving the people to do things that are useful for others decreases.

"Ideally", if 100% of most people's money comes from the helicopters, their own work doesn't really matter. It's up to them whether they work and we end up with communism – the ultimate social nightmare. People have whatever they want (up to the limits of the helicopter-rationed money) and they only work if they want. If you don't know what's wrong with such a setup, please let me know in the comments and I will immediately ban you (because beating you with a hammer could be viewed as a controversial response). Also, if such dangerous tools get out of control, a country may easily end up with hyperinflation of Robert Mugabe's style.

Deflation isn't a clear enemy

OK, let me start with the "enemies" of Brad DeLong. First, it's the deflation (or low inflation). He and others say that it's terrible. An emotional argument is that there was deflation during the Great Depression of the 1930s so it must be bad. However, correlation let alone coincidence doesn't imply causation. The deflation of the 1930s was mostly a consequence, not a cause, of the Great Depression. You know, the actual event that started the depression was a collapse of the stock market – the fact that some investors (who are a very similar group to savers!) got poor.

Moreover, in the whole last third of the 19th century, the U.S. has witnessed a persistent deflation. It happened to be the period of an unbelievable economic growth. The growth around 10% lasted for many years. Deflation obviously wasn't a problem. The prices were dropping because people could indeed produce things increasingly easily. It was ever easier to get a growing amount of commodities, to do something with them, the amount of human work needed to process something was decreasing because machines and better production methods were spreading, and so on. There was some "obviously real technological improvement" that allowed the prices to drop naturally. People knew that the prices would generally go down – a part of the progress as they witnessed it – but it didn't prevent them from buying things because they felt rich and optimistic, anyway. I am confident that such an "obviously real technological improvement" exists even today and if the deflation were up to 1% or 2%, it just couldn't have any truly bad consequences for the economy. Some pundits claim that deflation is a defining sign of a healthy economy. I am not sure I would get this far but I have some sympathy for such claims, anyway.

My actual viewpoint is much more neutral. For me, inflation or deflation is just a particular time-dependent choice of "units" of wealth, and the choice of units is basically unphysical. The U.S. dollar is 112 yens or so. But no one claims that it means that this ratio implies that Japan is by two orders of magnitude "worse" in some sense. After all, 1 yen is close to 1 U.S. cent. It's just a word, whether it's a yen or a cent, and it just can't matter much.

But even if the currency is changing the value as a function of time, e.g. because of inflation or deflation, it doesn't mean much. If the inflation rate were 20% a year but you would be getting 20% in interests, your "real savings" would be staying constant. For every inflation (or deflation) rate, it is possible to increase the interest rates and all other rates accordingly and nothing "important" is really changing about the economy, at least as long as some rates that may matter (e.g. interest rates for regular savers) are kept from dropping to the negative territory.

So I don't buy the claims that inflation below 2% is pathological. Inflation around 0% or even deflation up to 2% could be perfectly OK, as far as I can say. An actual argument I find legitimate is that numerous central banks have vowed to keep the inflation rate near 2%. If they have promised such a thing, they should better try to do something to fulfill the promise because that promise has been used in the financial planning of many companies and people and a sloppy monetary policy that disrupts the corporate and individual planning is harmful. A policy that violates the promise is good news for 50% of the people but bad news for the other 50% – and the latter may find themselves in serious trouble that could ultimately affect the "lucky" 50%, too. That's why central banks are supposed to fight against this kind of instability.

Savings glut isn't a real enemy

The second big enemy mentioned by people such as Brad DeLong is the "global savings glut". People save too much, they argue. Implicitly, they also say that people borrow too little – because it's only the difference between "saving and borrowing" that influences the aggregate demand. Well, give me a break. Relatively to the savings, the debt is probably too high. It's true for governments, individual consumers, and others. We periodically face regional or global "crises" that are ignited by the risk that someone – or a country – will go bust. Many types of countries' debt and other kinds of debt are borderline unsustainable. I surely don't think that the policies should deliberately encourage the debt levels to grow even further.

But let's look at the savings. Do people save too much money? Some people surely do but personalities and attitudes are different. The tendency to save is the particular idiosyncrasy of some people who think that they just don't want to be in the skin of others – or other countries – that are in debt and lose some of their freedom because of that. Is the total inclination of the mankind to save too high? I don't think so.

There exists an even more important assumption I find problematic, however. DeLong and even much more mainstream economists automatically assume that by reducing the real interest rates in some way, they force the savers to learn to spend the money. The logic is that "the saving is made unattractive, so the savers are reeducated to become spenders". I don't really believe that at all. The truth is that lots of savers are basically forced to become even more dogmatic about their savings (e.g. after 2015 in which "all forms of savings" were losing their value) because they feel that their savings are going down in the real terms (or not going up sufficiently quickly) so they are under pressure and they need to reduce their consumption in order to compensate for the decrease of their wealth.

The opinion of a person whether he "needs to have a financial buffer" is basically predetermined (by his experience and psychology etc.) and central banks can't change it much. Central banks may only influence the way that the saver finds "optimal" in his predetermined goal to save or, on the contrary, borrow.

I believe that this argument is much more correct than the opposite argument that "by crippling people's savings, you will encourage them to spend money recklessly" – and it is much more correct for most savers in the world. The actual causal relationship is the opposite one than people typically assume. In countries such as the U.S., my argument isn't too important because the borrowers who are in debt are much more important than the savers and obviously, you encourage people to borrow more if the interest rates go down. But in countries such as Czechia (which has inherited the conservative and thrifty attitude to the money from Austria-Hungary – even though the amount of consumer debt was rising in recent years), the savers arguably still prevail over the borrowers. And I think that if you make the life easier for the savers, by making their real interests (or returns) higher, you will encourage them to spend. In particular, if a saver got sufficiently certain (or inclined to believe) that he will basically get some $20,000 every year almost for free, be sure that he will start to enjoy his life and spend the money. So I am confident that a central bank policy meant to increase the inflation rate should actually try to give the savers – and investors – some wealth "easily". There's also the point that if the holders of savings accounts are "punished" (although they have made no sin), this automatically helps the value of other types of investments (stocks, real estate, debt etc.). If you own $$X$$ percent of the "world's wealth", whatever its definition is, the sum of $$X$$ over all people just can't be changed because it is 100 by definition. In this particular sense, all efforts to change the "truly aggregate" wealth of all the savers is tautologically an oxymoron. Principles that decide about the wisdom of QE-like policies OK, let me now assume that the central banks really should try to return the inflation rate up to 2% – because they have promised it. And they have already adopted all sorts of policies to increase the money supply. How should it be done? What are the fundamental principles that should decide what is a good policy (e.g. details about a quantitative easing) and what is a bad policy? Here are the principles: 1. A central bank is allowed to print the money so in principle, it's straightforward for it to increase the money supply and the inflation rate. Because it's so straightforward, every central bank should be aware of its responsibility because an intervention can have consequences for all parts of the markets and the economy. 2. The central bank should keep some "rates", e.g. the inflation rate, near levels that were promised and that were used by many users of the currency to do their financial planning. Certain "rates" are better off when they're stabilized. 3. When some of these rates are stabilized, the central bank should still think about the possible destabilizing influences that the policies may have on other, perhaps more important quantities, even in the long run. Hyperinflation, the spiraling debt, or sovereign defaults are just extreme examples. Any inability of the central bank to fulfill its promise is bad, too. 4. There are lots of "rates", such as the change of a price ratio in a year, that must be left to the free markets. Central banks should avoid "micromanaging" and interventions into the small parts of the economy where the invisible hand of the free market shows its ability to optimize the selection and allocation of the goods and the capital. 5. When policies are changed, there should be a clear desire to preserve some continuity of the price ratios (but not necessarily of their time derivatives) i.e. not to create one-time winners and losers. One has to be careful. Needless to say, Stalinist aßholes such as Brad DeLong don't give a damn about their possibility to screw things up. If they were central bankers, they would feel "entitled" to distort all the prices, manipulate everything, harm almost everyone – and their ideological justification that they have helped a bunch of particular privileged losers would be a sufficient excuse for them to get rid of the guilt for all the bad consequences of their sick policies. But a wise central bank should have some respect for the things that work – for the justice; for the optimal correlation between the work and genuine contributions to the society on one side, and wealth on the other side; for the invisible hand of the markets that optimally decides about many prices whose values are needed to decide about "what's the best way to do things". OK, once you agree about the moral and effectiveness-based dimension of the problem – that unaccountable insensitive aßholes such as DeLong just shouldn't be allowed to influence things – you are ready to discuss the technical issues. My attitude is that central banks have some power and they influence things in one way or another. Their behavior may look more active or less active, more creative or less creative, more intrusive or less intrusive. But these are just appearances and central banks always influence some things – even if they seemingly do "nothing". They should better influence the "macro" things only, instead of trying to distort things at the microscopic level. Sane people will hopefully agree that a central bank shouldn't try to force people to drink Pepsi instead of Coke and so on. What the central banks should do? They can do a lot. As we have seen, they can print the money and buy all kinds of assets. The government bonds have been the most frequent ones (it's an asset swapping that doesn't do too much) but the central banks have already bought lots of corporate debt (the European Central Bank has included the corporate bonds among the things it's planning to buy or already buying). The central banks could also buy foreign currencies (like the Czech National Bank which buys lots of euros while it enforces the CZK 27/EUR level); stocks; real estate; land; commodities; food; and even patents, copyrights, whatever you like. In some sense, all the things that a central bank buys may be said to "partly backup the fiat money". They protect the value of the fiat money at least to the extent that these assets of the central bank may be sold to stop (or decelerate or delay) the drop of the currency if one arrives. Every purchase like that carries the risk that the central bank would be distorting a part of the markets and it's a bad thing. How do you separate the "good purchases" from the "bad ones"? Obviously, there can't exist any "totally sharp" border separating these two groups. The legitimacy of some program is a continuous quantity. However, there exist mostly relative prices that must be left to the free markets because it has been shown that the free markets change the prices or price ratios of these things for good reasons, and they are actually very helpful for the optimal allocation of goods and resources. But there exist "overall" prices, indices, inflation rates etc. that simply don't play this role and almost any instability in these quantities may be viewed as an unhelpful external random oscillation if not a macroeconomic disease. We may consider the inflation rate that is "extremely far from 2%" to be such a sickness which is why central banks may target the inflation rate (although they may also target the nominal GDP/GNP rate etc.). However, I think that it's right to say that there are many other quantities whose volatility is "not helpful". In particular, the Great Depression started in 1929 when the Dow Jones index lost most of its value. For this reason, I am willing to argue that similar highly inclusive stock market indices represent quantities similar to the "general price level" (whose logarithmic time derivative is the inflation rate) and it may be a good idea for central banks to "slightly" regulate them. What I have in mind is that in order to "prevent new Great Depressions caused by the mass hysteria", central banks could promise to keep the country's "main" stock index from dropping by more than 20% in 365 days. Central banks may easily fulfill this promise. If the stocks drop by more than 19.9%, they print a sufficient amount of new money and start to buy stocks! With this commitment, I would find it natural to have some kind of an opposite commitment as well. When the stock market index goes up by more than 30% in 365 days, the central bank is obliged to start to sell the stocks in its portfolio and keep the year-on-year growth below 30%. You would be correct if you said that by adopting many such new policies and algorithms, the central bank would gradually be redefining the "definition of the money". As I said, a central bank should avoid interventions to the "relative prices". So it should buy and sell all these stocks "fairly". I am not 100% sure what's the best quantitative definition of "fairly". I can imagine that the central bank buys the number of each stock every day that is a fixed coefficient $$K$$ times the average daily volume traded on the stock exchange every day in the previous year. There may exist other quantities to which the "right percentage" should be pegged. I think that if you vaguely okay my proposal, it's an important question to decide which of the algorithms is the "least intrusive" one. The minimization of the intrusion should also try to keep the price ratios unchanged relatively to this pre-intervention era. But with this disclaimer – that the bank should maximally avoid the distortions of the relative prices that are better left to the free markets – it seems utterly reasonable to me that the central banks could start buying stocks as, roughly speaking, represented in the stock indices. This is an effective way to send more cash to the real economy – through the stockholders – and increase the price levels in the "sensible" way. My guess is that if the central banks decided that my proposal is a great idea and they would double the stock market indices across the world by the end of April 2016 ;-), it would be great news for the world economy. Such an increase of wealth wouldn't produce too much inflation (measured by the prices of everyday products) simply because the stockholders usually have enough money to afford ordinary everyday products, anyway. However, because the stock holders are more likely to be employers, this central-bank-made increase of their wealth would probably lead to increased hiring and investments of many sort. Slowly, the wages would also grow and the increased wealth of the ordinary people would drive inflation, but indirectly. It would be a good thing. Yes, I think that if the central banks artificially turned the top 10 wealthiest people in the world into dollar trillionaires by 2018, the world would be better off, too! Lots of new jobs designed to build Moonlanding rockets and colliders for trillionaires and some multibillionaires would be born, too. The concentration of capital is something that allows the technological and social progress. It is tightly correlated to the sophistication of the society. The more primitive and poorer the society was in the distant past, the more egalitarian it was. Many leftwingers don't want to see this obvious truth – which conflicts with a dogma of theirs – but that blindness can't change the fact that the truth is true. Such production of very rich people is much better thing than to drop the money to poor neighborhoods from the helicopters. At the end, if policymakers agree that the money should be dropped, the primary question is whether they should be dropped to "mostly poor people" (or borrowers) or "mostly wealthy people" (or savers). I am confident that the latter possibility is much more helpful for the economy. If I simplify just a little bit, the former possibility more or less means that you increase the prices of alcohol and cigarettes (and their production) because most of the people who acquire the helicopter banknotes will buy these two products quickly! This shouldn't be a holy goal you should get enthusiastic about. The trickle-down helicopter money for the investors do much more useful work and they are much more consistent with the rules of the market because they're basically reversible asset swaps. They are much more likely to improve the economy and the society in the long run because the richer people – and people who have no debt – are much more likely to care about the long-term goals and to use the money (=the extra power and freedom that comes with them) wisely. The central banks could buy stock indices and they could promise to partially stabilize overall stock market indices, and perhaps (to a lesser extent) even some other indices, exchange rates, and other financial variables. I believe that the volatility in the most inclusive stock market indices – and perhaps even a more specialized indices – is basically a bad thing, not an example of the invisible hand of the free markets that is doing a helpful work. You could see that the world worked when the euro was$1.45 but also when it was \$1.05 – nothing "truly material" has changed by 30% during the year which indicates that these big swings don't play a very useful role and could be partly regulated away (e.g. by an agreement of central banks to keep the maximum change at 15% a year). This volatility is only good (or, sometimes, dangerous) for speculators who don't do much useful work, anyway. In a civilized economic system, such quantities could be more (partially) stabilized by some mechanisms. It may be true for some exchange rates, too.

I believe that a world that wouldn't allow something as disgusting as a new wealthy George Soros to be born would be a better world. Much more generally, I think that many monetary or "arbitrary" macroeconomic variables – including the indices that could be called the overall fear index in the markets – should be kept more or less stable.

To summarize, I believe that the central banks and macroeconomists should be aware of the principles – it's great to stabilize things whose volatility is "almost purely harmful" while it's always bad to distort the relative prices where the free market sets the right prices for "very useful reasons". Economists should be spending more time with the question which directions in the "space of logarithms of prices" are better left to the free markets (the volatility has very good reasons if it's there) and which are better to be regulated (volatility is a psychologically or speculatively driven nonsense that has no other useful role). And once this is understood, the central banks could establish more resilient financial systems – systems where price ratios that have very rational reasons to move will move; and price ratios and rates that were just wildly fluctuating for no good reason (and because of mass hysteria) will be stabilized. At the end, because the oil price may be quickly changed by small changes to the supply, I do think that the oil price should be partly stabilized, too (supply cuts/increases should be mandatory when the price changes by more than X percent in a year etc.).

Meanwhile, if the goal is to increase the inflation rates back to the 2% level, it should be done by asset swaps, not by uncivilized procedures such as helicopter drops or subsidies for the poorest ones, random people, or debtors. All responsible people feel that "there is something dangerous about the central banks' printing and interventions". But if they think a little bit more carefully, they may see that most of the real danger may be basically eliminated if the interventions are done in a sensitive way. And that's the way in which such interventions should be done.