Deflation vs. inflation: which is it going to be? We've looked at what the terms mean. We've looked at the mechanisms that bring about inflation and monetary statistics to describe the money supply. Now its time to look at the reverse process: deflation. We will also look at a few phenomena that have no influence on the money supply, but are sometimes blamed for one or the other.
As you might have already guessed, deflation occurs principally through the reversal of the processes that lead to inflation. It can also be accomplished through a few other processes, which will also be discussed.
Sale of Assets Just as when the FED buys assets money is created, when it sells assets money is destroyed. Because of these twin properties, I like to think of the FED as a something like an infinite capacity black-box: when assets go in, money comes out; when assets come out, money goes in. That's the easiest description I can think of.
In a more detailed analysis, the FED will sell assets into the market, withdrawing money from the deposit accounts of buyers. The FRN liability (Federal Reserve Notes, or US dollars for those of you who forgot) is redeemed and retired by the FED as it absorbs the FRN into its vaults and out of the banking system, and the asset passes from the vaults of the FED back into the market. The asset and liability columns of the T-account are wiped clean:
In the course of this process, bank reserves are reduced as those FRN leave the banking system. If the banks are comfortable with their reserves and reserves are above the legal requirement, nothing more need occur. In our example, the bank is in trouble as it has no reserves and will have to sell other assets to meet legal reserve requirements. Or go to jail. If banks choose to increase reserves for whatever reason, the process continues. By selling their own assets or calling in loans, banks increase money held in reserve, clear their T-accounts of assets and contract the money supply further:
Our example is a little extreme for two reasons. First, I am feeling lazy and did not want to draw out a whole new T-account chart. I just cut and pasted it. Normally, it would be a little strange for a bank to sell to its own depositor, but I suppose it might happen. But it does illustrate a good point. Selling assets also increases the reserve ratio by reducing liabilities, e.g. deposits, e.g. the money supply, not simply increasing reserves. It is an unwinding of the fractional reserve process. I could have had the bank sell to another bank or another private entity to simply increase its reserves. But that would have placed a strain on the buying bank, which would have lost reserves and would likely have had to shuffle things around even more. And I just can't stand the thought of having to draw out another and another of these durn things to explain all that.
The net effect overall from the banks' point of view is a transfer of assets out of the banking system and a reduction of bank deposit liabilities. From our point of view, it is a reduction in the money supply. Bonus question: what is the net effect when fractional reserve banking is used by banks to buy assets, then when the banks get in trouble and are unable to pay back depositors, the FED prints money to bail them out instead of forcing them to sell their assets to raise the money? Hmmm? Anyway, the point is that selling assets effectively reduces the fractional reserve multiplier, and the money supply. That constitutes the "desirable" deflationary routes engineered by the FED for control of the money supply, and autonomous contraction initiated by banks themselves. Incidentally, from a graph of the AMB, it is clear that the FED rarely deflates through the sale of assets. M1 and M2 indicate that the banks themselves are more likely to contract the money supply on their own, and at those times not by very much. For the most part, the monetary base is continually increasing, with short periods of stagnation. S0, there you have it, the two major forces of deflation: banks allocating more money towards reserves, e.g. decreased lending/selloff of assets, and the sale of assets by the FED. The exact opposite of inflation.
Other Deflationary Forces
The other routes for deflation are not nearly as important. First, depositors could withdraw currency en masse and hold it as cash outside of the banking system. This is deflationary because money held in physical cash is not on deposit with the banking system. It is held in deposit in your pocket, and cannot serve as reserves for the fractional reserve process while it is sitting in there. If enough cash is withdrawn, it will force banks to begin selling assets and call in loans in order to meet reserve requirements, or it will force the FED to purchase assets to provide banks with new reserves.
Of course, eventually the money that is withdrawn is likely to be redeposited and the effect reverses. Two exceptions exist. The first is that the withdrawn money is destroyed, as in, put into a pile and set on fire. Obviously, this is rare. As angry as some of us may get over inflation, we don't generally do this.
The second is not so rare. The money can be held permanently by private persons outside the banking system. This is most frequently the case for money that leaves the country. Many foreigners choose to hold paper US dollars as savings in preference to their local currencies as the purchasing power of the US dollar has tended to hold up better in the recent past, despite the inflation engineered by the FED. These dollars circulate in "bankless" black markets, or are simply stored up for a rainy day.
Central banks can (and do!) also hold US dollars as reserves, though they usually prefer Treasury debt. These dollars "back" their own currencies, just as our currency is "backed" by the Treasury debt and other assets held on the FED's balance sheet. By buying and holding dollars, foreign central banks create more of their own currency and reduce the quantity of dollars in circulation, artificially propping up the value of US dollars and distorting trade. But that is a long and involved topic of its own best left for another time... Imagine: there are banking systems out there even more screwed up than ours! In any event, this effect tends to tamp down the inflationary effect of printing ever more dollars, but could come back to haunt holders of US dollars in the event of a major devaluation. But I suppose that in a major devaluation this will be the least of our worries.
Bank Failures and the FDIC
At one time, bank failures also resulted in deflation of the money supply. Bank failure occurs when financial strains cause the value of a bank's assets to be substantially less than the liabilities imposed by the bank's expenses, deposits, or depositors attempting to withdraw their funds all at once, e.g. a bank run. Over the past year or so, a few banks have failed almost every week. If you check Yahoo!Finance late on Friday afternoons, after markets have closed, you can usually read about them in the news stories. The FED closes them when markets are unable to respond to prevent panic. It's been happening like clockwork for some time, though most people pay little attention.
Since the introduction of the FDIC, bank runs have become virtually nonexistent, but at one time they were frequent. Note that bank failure is not the same as bankruptcy. Technically, banks are always bankrupt, in that they cannot possibly honor all their contracts simultaneously. They only stay in business because not all contracts are enforced, not to mention because the FED and the government cover for them when they get into trouble. This is different from a businessman who owes more money than his business is worth. The businessman is not obligated to pay his debts all at once; his debts are amortized, and as long as he can make his payments, he is still in business.
In contrast, a bank is required to refund all of a depositor's money at any time. All the depositor has to do is ask. This is a basic problem of time mismatch: borrowing short and lending long. The carry trade. Whatever you want to call it, depositors essentially act as ultra-short term lenders or creditors, while the bank is busy making long term investments. The bank can't possibly fulfill the terms of its agreements, if they are enforced. The banks are fully aware of this discrepancy in expectations, and in many ways it is a sort of fraud.
This is precisely why deposits are listed under the heading liabilities. The bank is obligated to pay them back. In the old days, when a bank run would occur, word got out and depositors began withdrawing money until there was none left, then the bank was forced into liquidating its assets and repaying what deposit claims remained with the funds raised through asset sales. Naturally, there was not enough, and some depositors lost money into thin air. Money was destroyed in the process, and M1 or M2 would have reflected this loss as deposits evaporated and the money supply contracted. Hence, lost deposits due to bank failures results in deflation. With FDIC protection, accounts are insured up to $250,000. Most people are smart enough to spread out deposits so that they get full protection, but theoretically if a deposit was over this amount, the surplus would evaporate in the event of a failure.
You can simply accept that "deposits are protected" but it is instructive to look at the actual transactions to understand how money is conserved by this process. The FDIC does not actually have any money. It has assets in the form of Treasury debt, much like the FED itself. When the FED takes control of a failing bank, it begins the process of liquidating assets and paying off depositors. The remaining funds that must be raised are obtained by selling the assets held by the FDIC. These assets are sold to other banks, which buy them with money from reserves, initiating fractional reserve multiplication. So the money "lost" in the failure is made up for by fractional reserve multiplication. Pretty simple, really, but it took me awhile to get it.
The FDIC was created in 1933. Prior to 1933, bank failure resulted in deflation, but since that time it has not. So long as Congress continues to authorize funds for this agency, it will not. There is no need to fear a catastrophic deflationary spiral as a result of bank failures, which some claim occurred in the early years of the first Great Depression. Of course, it is still a political question, since the FDIC has already pretty well exhausted its asset base and Congress could theoretically refuse to provide any more funds, but I think it is a rather safe bet that Congress would not do that.
Other Notable Absences: Deficit Spending, Default, and Falling Asset Prices
A few readers may not see their pet deflationary or inflationary forces at work here. Of course, I have missed some, as this is not exactly an exhaustive work. On the other hand, there are many events which have been attributed as inflationary or deflationary but in reality are not. I'll go through a few here.
Contrary to popular belief, deficit spending is not inflationary. Deficit spending is funded by issuing Treasury debt certificates. These certificates are bought by private investors and the money is transferred to the government and spent into the economy at large, passing back into private hands. No money is created in the process.
Deficit spending by the government is no different from the issuing of corporate bonds. The result is private transfers of money, not creation of money, and not inflation. Banks may buy the debt with depositor funds, yes, and in this case it results in multiplication by fractional reserve accounting. But the inflation is a result of the accounting, not the issuing of the debt itself. The FED may buy the debt directly, resulting in inflation of the monetary base, but again, that is as a result of the bank buying the asset, not the debt itself.
One way that deficit spending can get politically intertwined with inflation is that a large deficit can create a political incentive for the FED to buy the debt, increasing the monetary base in the process. Many, if not all, central banks engage in this nefarious practice regularly. By having the central bank buy debt, the government can obtain loans at interest rates far below market rates since the central bank supplies artificial demand for the debt. And when debts begin to pile up, it can be tempting to have the central bank simply buy up all of the debt and forgive interest payments owed by the government. This results in large-scale inflation for holders of the currency and repayment of debt in currency of depreciating value, effectively giving the governments creditors the shaft and allowing the government to default on its debts without legal bankruptcy. Nice of them, huh? I guess arbitrary authority is a good gig if you can get it...
This is called debt monetization, and though it will destroy a nations credit rating and its economy, it becomes a very real possibility as interest payments begin to absorb a politically unacceptable fraction of the budget, or fiscal deficits begin absorbing politically unacceptable fractions of GDP. The former will generally lead to the latter. Many nations are approaching these levels of debt, in particular the US and Japan, two of the largest economies in the world. A rise in interest rates to normal levels after years of suppression by central bank policy could effectively render the debt unpayable, especially for Japan where debt approaches 200% of GDP.
But, as I said, fiscal deficits in and of themselves do not cause inflation. It becomes inflation when central banks like the FED fund the deficit by buying debt certificates. Whether or not the central bank chooses to do so is a political question, not an economic question, though it does have economic consequences.
Occasionally, you will encounter the belief that default on a loan destroys money and results in deflation. You are especially likely to believe it if it is your money! This is not the case, however. The lender will not be paid back, and the price of his asset (the bond) goes to zero. But the money he lent to the borrower is still in the economy, whether or not he ever sees it again. Whether or not loans are repaid has no effect on the money supply.
A fall in asset prices, like a stock market crash or the popping of a housing bubble, does not result in deflation either. The price of the asset in question simply falls. Your investment account or 401(k) does not "have less money in it." It is "worth less" (or just "worthless," as the case may be!) But the same amount of money is still floating around out there in the economy. A brokerage account contains assets, not necessarily money. A rising stock market does not create money and a falling stock market does not destroy it.
Monetary inflation occurs principally through the purchase of assets by the FED and through fractional reserve banking. Fractional reserve accounting increases the money supply by counting the same money twice, once as deposits and a second time as money lent out. By lending money out which gets re-deposited into the banking system, fractional reserve banking can increase the money supply by up to ten-times reserves, which are the initial deposits provided by the FED through purchase of assets.
Deflation occurs principally through the opposite mechanisms: sale of assets by the FED and banks choosing to increase their cash reserve holdings by selling assets and calling in loans. A few other avenues of deflation exist, but these are more limited in scope. Deficit spending, market crashes, defaults and bank failures do not result in changes to the money supply. These forces, however, can exert political pressure for central banks like the FED to create inflation in order to "paper over" these problems and allow debtors to legally shirk their obligations.
Next up: how the FED influences interest rates!