What do central banks do?

The central bank (CB) is the monopoly producer of high-powered money (HPM), which makes up the monetary base. HPM comes in two flavors: (1) physical currency--paper notes, coins, etc.; (2) electronic bank reserves. In a typical developed economy, commercial banks must, by law, deposit some fraction of their customers' deposits into electronic vaults supervised by the CB. The funds put towards meeting this mandate constitute a bank's 'required reserves'. Any extra funds the bank deposits constitute its 'excess reserves'. The CB is the monopoly producer of HPM for two reasons: (1) it is the only entity legally permitted to create new physical currency; (2) it is the only entity legally permitted to electronically credit participating banks' reserves.

The CB is typically free to produce however much HPM it chooses, in either form. Additionally, HPM's cost of production is, approximately, zero. Consequently, the CB controls the supply of HPM in the relevant currency zone.

Why does HPM matter? The prices of goods and services are quoted in units of HPM--HPM is therefore the 'unit of account'. HPM is also what consumers use to buy goods and services--HPM is therefore the 'medium of exchange'. In equilibrium, the value of HPM is determined by the supply of, and demand for, HPM. Hence, in the long run the CB, through its management of the supply of HPM (its 'monetary policy'), determines the value of HPM, which is the flipside of the price level--the average price of goods and services. The higher the value of HPM, the fewer the units of HPM necessary to purchase a given bundle of goods and services. The lower the value of HPM, the greater the units of HPM necessary to purchase said bundle. As a result, monetary policy, executed by the CB, determines the price level, and therefore the rate of inflation (the rate of growth in the price level), in the long run.

If prices are perfectly flexible, then the economy equilibrates instantaneously, which means the CB determines the price level and the rate of inflation in the short run, too. Moreover, its policies are otherwise irrelevant to the evolution of the economy. Economists inclined towards a flexible-price view of the economy, therefore, believe that the sole objective of monetary policy ought to be 'price stability', usually defined to be a low and stable rate of inflation. If prices respond to shocks only sluggishly (that is, if prices are 'sticky'), however, then the economy takes time to equilibrate, which means that the CB plays a bigger role in the short-run evolution of the economy. Economists inclined towards a sticky-price view of the economy, therefore, believe that monetary policy ought to concern itself with more than mere price stability.

Suppose, for example, that the demand for HPM jumps (for whatever reason). This puts upward pressure on the value of HPM, meaning downward pressure on the price level. If prices are sticky, however, then many prices will remain too high in the face of this pressure. When the price of a good or service is too high, producers have the capacity to produce more than consumers want to consume. Producers react to this demand shortfall by contracting their output, rendering some of their inputs (e.g., labor) redundant. If wages are sticky, too, then redundant workers will continue to seek employment where there is none. As a consequence, output falls, while unemployment rises. If, in response, the CB expands the supply of HPM, this puts downward pressure on the value of money, meaning upward pressure on the price level. This offsets the downward pressure on prices, restoring the economy to equilibrium. As a consequence, output rises, while unemployment falls. Note that, in the process, the price level more or less stays put--monetary policy is impacting the economy without much impact upon price stability.

Considerations of such possibilities lead sticky-price economists to pin the blame for business cycles on the CB. When the CB does not provide the economy with enough HPM, it causes a recession. When it provides too much HPM, it causes high and/or unstable inflation. Monetary policy, therefore, has as its objective not only price stability in the long run, but also maximum output/employment in the short run.

What, then, do interest rates have to do with monetary policy? Interest is the price of a loan. A higher interest rate causes savers to save more, borrowers to borrow less. Savers make up the difference by building up their HPM reserves, expanding the demand for HPM, which (other things being equal) causes a recession. A lower interest rate causes savers to save less, borrowers to borrow more. Borrowers make up the difference by drawing down their HPM reserves, expanding the supply of HPM, which (other things being equal) stokes inflation. Thus, the stance of monetary policy may be equivalently characterized by either the supply of HPM, or a target for a benchmark interest rate. (The CB usually has no reason to interfere with the pricing of risk, so it typically manages a benchmark interest rate with reference to which financial markets fix other interest rates.) How, though, does the CB move the market rate of interest in line with its target?

An open market operation (OMO) is a transaction wherein the CB buys or sells assets in the marketplace, by drawing down or building up its HPM reserves. Since the CB can, in principle, expand the supply of HPM without limit, it can, in principle, buy or sell whatever quantity of assets is necessary to move the market rate of interest in line with its target. Because it can do this, however, it need not do this. If the CB declares a target for its policy rate (the benchmark interest rate it manages), market participants understand that there is no point doing battle with the CB. The CB always has more HPM than you, by design. Consequently, communication is usually enough to move interest rates towards the target, though the CB often engages in medium-scale OMOs to show its resolve.

The usual business of the CB, therefore, is to publicly set its target for the policy rate in a such a way as to maximize output/employment, while maintaining price stability over the long run. Sometimes, however, managing the policy rate isn't enough for the CB to fulfill its dual mandate. More on this issue to come...

Misconceptions concerning inflation, Pt. II

Misconception #4: Inflation is the very same thing as printing money. Define inflation however you like--be my guest. Just note that if inflation is defined to be the printing of money, then the dominant theory of inflation, the quantity theory of money (QTM), becomes a tautology. The QTM says that x% money growth, in the long run, causes x% inflation. If inflation is the very same thing as printing money, then the QTM says that inflation leads to inflation in the long run. No shit. Let's call rises in the price level 'schminflation', then, so we can carry on.

Misconception #5: Inflation causes bubbles, which cause busts. I should begin by observing that no one has a definition of "bubble" that satisfies anyone else. Note that just as the US housing bubble was supposedly developing, home prices were appreciating in many other countries, too. While US prices eventually came crashing down, prices in many other countries (e.g., Canada) have kept on rising. Have these countries figured out ways to keep their bubbles from bursting, or should we conclude that the diagnosis of bubbles ought to be left to Captain Hindsight? Prices go up, then they go down, then up, then down, ... What insight is gained when we conclude that some one upturn followed by a downturn was a bubble, when no one can systematically identify bubbles ex ante?

In any event, below is the inflation rate during the 2000s:



And here is the Case-Shiller home price index during that same period:



Which came first, the housing bubble or the uptick in inflation? And which was first to reverse course? As for bubbles causing busts, note that the housing market began to fall apart long before the broader economy followed suit. Indeed, the collapse in the housing market didn't even accelerate once the economy went into the tank. By contrast, inflation lingered at 2.0-2.5% per year before plunging in concert with the broader economy. A prescient paper by Ben Bernanke (with a coauthor) contends that the responsibility of monetary policy is not to identify bubbles in order to burst them, but rather to keep the broader economy stable as it absorbs the shock. The economy didn't tank because the housing bubble burst; it collapsed because the Fed failed to keep AD on target. Had the Fed been aggressive enough in its response, the housing bubble would've ended with a whimper, not a bang. And it wouldn't make a difference whether we thought the movement in home prices was a bubble or not. Inflation would seem, correctly, to have nothing to do with any of it.

Misconceptions concerning inflation

We want a positive rate of inflation in order to keep the unemployment rate at tolerably low levels. We want a low rate of inflation in order to minimize relative price distortions due to price stickiness. Most importantly, we want a stable rate of inflation in order to facilitate intertemporal transactions (for example, almost every financial transaction). Occasionally, we may want a higher than normal inflation rate (now is one of those times) in order to lower real interest rates when nominal interest rates are stuck at zero. Normally, however, a low, positive, and stable rate of inflation is what we seek, and ever since Paul Volcker broke the back of unstable, double-digit inflation during the early '80s, the central bank of the United States, the Federal Reserve System, has more or less delivered on those inflation objectives. Why, then, do some people worry so much about inflation, and insist upon monetary regime change in order to contain an inflation problem that exists only in their own minds?

Misconception #1: Inflation hurts those who save. Recall the lending example in my previous post. If I naively lend you my savings, without taking inflation into account, then I will end up getting back less than I was expecting. If inflation is stable, however, I need only adjust the interest rate I charge you for the rate of inflation in order to get back exactly what I was expecting. In general, so long as inflation is stable, savers need not fear it, for they may simply demand higher interest rates to compensate them for it, which is what savers in fact do. (This is why high inflation is correlated with high interest rates, low inflation with low interest rates, etc.) The only savers that stable inflation hurts are those who save by hoarding cash (by stuffing it under their mattresses, for example). These people, however, have nobody to blame but themselves. Even in the context of perfect price stability (0% inflation), the rate of return on cash is inferior to equally safe alternatives (e.g., Treasury securities). And if you're really concerned about inflation, you can always buy TIPS (Treasury inflation-protected securities), which adjust Treasury yields for CPI inflation. The reality is even the most insecure people typically don't stuff a lot of cash under their mattresses; instead, they stick their cash in government-insured bank deposits. And banks, of course, adjust their interest rates for inflation. Savers have nothing to fear, therefore, besides unpredictable inflation, but everyone has a stake in keeping inflation predictable, not just the savers.

Misconception #2: Inflation raises people's cost of living. Inflation is any rise in the general level of prices. Hence, the more inflation we experience, the higher the price of gas, right? The higher the price of gas, the higher your cost of living, right? Not quite. Inflation does show up in higher gas prices, but it also shows up in higher wages. Much like savers adjust interest rates to compensate for inflation, workers adjust wages for the same reason. Even though gas prices are higher, so is your income, which means gas is just as affordable now as it was when gas prices were lower. The same reasoning applies to other commodities. Inflation raises the prices of the things you buy, but it also raises the amount of money you have with which to buy those things. It has nothing to do, therefore, with your cost of living. What does determine your cost of living? One word: scarcity. The less stuff there is, the more expensive it is to buy. If higher gas prices raise your cost of living, it is not because the Fed is printing too much money. Rather, it is because there's not enough gas out there to support the previously lower price of gas. Unfortunately, Dr. Bernanke is not known for his oil drilling skills, seeing as the only way he may lower gas prices without also lowering your wages is by drilling for more oil.

Misconception #3: The gold standard is the best recipe for keeping inflation in check. This is probably the nuttiest idea out there. The gold standard is a monetary regime in which the money supply is adjusted over time in order to keep the money-price of gold fixed. The idea is that gold's value is inherently stable, and thus the value of money would be thereby stabilized. Here's the problem: suppose miners in South Africa unexpectedly discover enormous reserves of gold. Such a discovery would sharply depress the goods-and-services-price of gold, but in order for the money-price of gold to remain fixed, the goods-and-services-price of money would have to fall sharply, too, generating a sudden burst of inflation. Sound stable to you? Or consider a historical example: just before the Great Depression, France began hoarding lots of gold, for reasons unknown to the rest of the world. This raised the goods-and-services price of gold dramatically, which under the gold standard meant that the good-and-services price of money had to rise dramatically, too. The result was that France imposed catastrophic deflation on every country on the gold standard, triggering the Great Depression. How do we know this? Countries like China and Spain, which were not on the gold standard, suffered only mild recessions, and essentially no deflation, because of the collapse of their trading partners. Every country on the gold standard plunged into depression, in the midst of historically rapid deflation. But here's the best part: the timing with which countries began to recover from the Great Depression is precisely the timing with which countries abandoned the gold standard. Here is US industrial production at the beginning of the Depression:



Guess what happened right at the trough in early 1933? FDR took the US off the international gold standard. For serious.

The last few decades show that determined central banks are perfectly capable of keeping inflation at low, positive, and stable rates, without surrendering monetary policy to the random forces at play in the gold market. Why, oh why, does anyone want to go back on the gold standard?

Let's talk about inflation, baby

Inflation is a concept that many people seem to struggle with. A brief overview may be helpful.

What is inflation? Inflation is any rise in the general level of prices. What, then, is the price level? It is the average price of goods and services produced by the economy. Suppose, for example, that the price of everything doubles over the course of a year. In that case, the average price of goods and services--that is, the price level--doubles, too. The rate of inflation, in turn, averages 100% per year.

What determines the rate of inflation? The price level is determined by the intersection of the aggregate demand (AD) schedule and the aggregate supply (AS) schedule. Thus, the rate of inflation is determined by the rates of change of AD and AS.

The value of money is determined by the intersection of the money demand schedule and the money supply schedule. The prices of goods and services are quoted in units of money. As a result, it makes sense to think of the price level as the money-price of goods and services, and the value of money as the goods-and-services-price of money. These variables are two sides of the same coin. A higher price level means that one needs more money to buy the same goods and services, implying a lesser value for money. A lower price level means that one needs less money to buy the same goods and services, implying a greater value for money. The price level is thus simultaneously determined by the intersection of the money demand schedule and money supply schedule. Consequently, the rate of inflation is simultaneously determined by the rates of change of money demand and money supply.

The economic function of inflation is to concurrently equilibrate the market for goods and services (AD and AS) and the market for money (money demand and money supply). Suppose, for illustration, that the government creates some extra money, with which it buys things. This expands the money supply, putting downward pressure on the value of money, which in turn puts upward pressure on the price level. The result, in equilibrium, is inflation. Another way of seeing this play out is as follows: when the government buys these things, it puts extra money in the hands of people who were previously satisfied with their money balances. Not needing to add to their money balances, they spend the money instead, expanding AD, putting upward pressure on the price level. Again, the equilibrium result is inflation. The reverse experiment, contracting the money supply, puts downward pressure on the price level, the equilibrium effect of which is deflation (the opposite of inflation).

Why does the rate of inflation matter? First, in the short run, prices are sticky. The higher the rate of inflation, the faster each price must adjust in order for relative prices to remain the same. Price stickiness means that some prices may not keep up with the rest, altering relative prices, causing resources to be misallocated.

Second, inflation instability undermines intertemporal transactions. Suppose I lend you $100, to be paid back in one year at a 10% per year interest rate. A year from now, you will pay me back $110. If, in the interim, the rate of inflation averages 20% per year, then $110 in next year's dollars is equivalent to $91.67 in this year's dollars. You'll pay me back less, in real terms, than I intended. If instead I charge you a 32% per year interest rate, then a year from now you'll pay me back $132, which is equivalent to $110 in this year's dollars. By modifying the interest rate that I charge, I may nullify the effects of inflation. I can only do this, however, if the rate of inflation is predictable. If it isn't, then I do not know how much I will be paid back, regardless of the interest rate I charge; hence, I will be discouraged from entering into this transaction in the first place.

Third, there is a short-run tradeoff between inflation and unemployment. More AD/money supply, in the face of sticky prices, boosts employment at the expense of higher inflation, while less AD/money supply contains inflation at the expense of lower employment.

Ergo, the dual mandate of macroeconomic policymaking: maximum employment (enough AD/money supply to fully employ the labor force) and stable prices (not so much AD/money supply as to produce high and/or unstable inflation). The question, then, is which regime is most conducive to fulfilling this dual mandate. I believe the answer is NGDP targeting, but either way the above is most everything you need to know about inflation. In my next post, I'll try to put to rest some misconceptions about inflation. (E.g., inflation is bad for savers, inflation raises people's cost of living, the gold standard is the best policy for managing inflation, etc.)

The Fed sets a nominal target

Well, it looks as if the FOMC read my post about the importance of setting a nominal target, as they have decided to adopt an inflation target of 2%. This is not my preferred policy, but it constitutes progress in my view. How so? Now that the Fed has a nominal target, curious minds will begin to wonder, "why 2%? why not 3%?", or "why inflation? why not nominal GDP growth?", or "why a growth target? why not a level/path target?", etc. Stick those in a blender, and suddenly the financial press will be wondering, "why not a 5% nominal GDP level/path target?". Once that question comes on the radar, the Fed will have some 'splainin to do.