In an interview, Larry Summers says,
The classic concern with respect to imbalances is that a situation develops where people are both trying to take money out of a country’s banks and trying to sell its currency. And the central bank can’t both add liquidity to help the banking system and the economy and reduce liquidity to maintain currency stability.
This was a new one on me.
I understand that if you raise the growth rate of the money supply, the currency depreciates. And I can see where banks that are suffering from a run might need reserves, and if the central bank adds reserves then this increases the money supply. But…
When people withdraw money from banks, isn’t that contractionary for the money supply? In the money and banking unit of freshman econ, if the public switches out of demand deposits and into currency, then banks have to cut back on lending, and the money supply goes down.
So I’m having a hard time worrying about the monetary expansionary effects of adding reserves, when what those reserves are doing is offsetting the contractionary effect of people pulling money out of banks.
Maybe the concern is what to do about the additional reserves once the bank panic stops and people are putting their money back into banks. But at that point, can’t the central bank sop up the extra, or just raise reserve requirements?
Larry, if you’re reading this, leave a comment and clarify. (This is sort of like the story Tom Wolfe tells in The Electric Kool-Aid Acid Test of Ken Kesey and the Merry Pranksters putting up a sign saying, “Welcome to the Beatles” in the hallucinatory hope that the Beatles would show up.)
Thanks to Greg Mankiw for the pointer.
The entire interview is recommended. Summers gives a number of rationales for having wise heads manage global economic problems. My views differ from his, but his views are well worth reading.
READER COMMENTS
Luke
Nov 1 2007 at 1:08pm
When people withdraw money from banks, isn’t that contractionary for the money supply?
No it is not. While the money has been withdrawn from the bank by the account holders, it is still in circulation.
John Thacker
Nov 1 2007 at 1:54pm
No it is not. While the money has been withdrawn from the bank by the account holders, it is still in circulation.
Luke, you either didn’t read or didn’t understand. As Prof. Kling says, banks have to cut down on lending when the money is withdrawn as well. Since banks only hold a small fraction of their outstanding loans in reserves, the money supply decreases, on net.
Incidentally, this is part of why I’ve never comletely understood the claimed multiplier effect for consumer spending, at least as opposed to saving. It seems like consumer saving would also cause a multiplier effect, due to lending.
ErikR
Nov 1 2007 at 2:06pm
The multiplier effect of bank reserves is obsolete and irrelevant. John Hussman provides a good explanation:
http://hussmanfunds.com/html/fedirrel.htm
“The point is simple. Commercial, industrial and consumer loans no longer have any link to bank reserves. Since 1995, the volume of such loans has exploded, while bank reserves have actually declined . Look at the one monetary aggregate that the Fed can directly control – the monetary base. Every bit of increase since January 1994 is accounted for by currency in circulation, not bank reserves.”
Bill Woolsey
Nov 1 2007 at 4:17pm
The tie is between checkable deposits and reserves. The process by which a reserve deficiency in the banking system impacts checkable deposits in the system is through changes in the lending.
Bank loans can expand while reserves shrink. That is because the banks can obtain the funds to make loans from sources other than checkable deposits.
Anyway, bank runs do result in a contraction of checkable deposits. _Ceteris paribus_ this involves a decrease in bank lending. But that could mean that bank loans expand less than they otherwise would.
While it is of course possible that many things could be happening at the same time as bank runs, but _certeris paribus_ lending will typically contract as well.
Anyway, I think that Kling is correct that bank runs would tend to stop a currency from depreciating on foreign exchange markets. The central bank might like this, but not like the fact that the runs are causing the banks to fail.
I think the traditional monetarist approach is to argue that it is best to let the currency depreciate and maintain domestic nominal stability. In other words, it is probably a good idea to offset the effects of bank runs on the money supply.
ErikR
Nov 1 2007 at 4:52pm
The thing is that the checkable deposits (and any multiplier effect thereof) represent a miniscule fraction of the liquidity in the system. Again, Hussman has more:
http://www.hussmanfunds.com/wmc/wmc070917.htm
http://www.hussmanfunds.com/wmc/wmc070924.htm
Danny L. McDaniel
Nov 1 2007 at 6:50pm
The problem as I see it is once upon a time in America a man had $5.00 in his pocket. He could use it to buy either gas for his car to get to and from work or purchase a gallon of milk for his family. Since this man didn’t get paid until Friday he was faced with a real dilemna. There was real reason for personal anger for his economic condition. That doesn’t happen any more because of plastic, credit cards. He can swipe his card and be good to go. But his anger is now gone while his economic condition worsen. In economics we often talk about delayed gratification. Today we should about delayed destruction.
Danny L. McDaniel
Lafayette, Indiana
Gary Rogers
Nov 1 2007 at 11:19pm
I would suggest that most of the comments, including Arnold’s original observation are micro-economic views of a macro-economic situation. With the amount of money in circulation, convertibility of currencies, modern financial instruments and the large segment of the world economy that now takes place outside the United States, we have no way to know what the money supply is. I think it is safe to say, though, that converting demand deposits to currency no longer matters.
What I do see is the dollar continuing to slide against both other currencies and commodities. I don’t care what the Federal Reserve says; when the dollar is losing its value we are experiencing inflation. Domestic prices may not show it yet due to the inertia created by years of stable prices, but this is inflation and we will soon be seeing it in full force. What worries me even more, though, is that we have something like 5 trillion dollars held by Middle Eastern governments, another 1.5 trillion held by the Chinese and who knows where else our dollars have landed. It doesn’t take an economist to figure out that if you are holding depreciating dollars you are going to start selling. Once everyone starts selling, the only way to stop the run is to significantly hike interest rates, which puts us one step forward then about ten steps backward. This is why I think the Fed made a big mistake and sent the wrong signal with its latest rate cut.
On the other hand, we still have a serious problem dealing with the 900 billion dollar sub-prime mortgage situation and have another looming problem with an additional 900 billion dollars of deteriorating credit card debt that has also been sliced up and sold the same way mortgages are repackaged. What we have is 1970’s style stagflation. These are the imbalances to which Larry Summers was referring, and I agree with him. It is going to take some real leadership to handle what is coming.
The question is how we got where we are. I suggest that the answer is also in the article, which mentions the productivity gains in the 1990s that allowed us to maintain both low unemployment and low inflation. Contrast this with the zero productivity created by war spending and we have the opposite situation. Just like the Viet Nam era spending caused stagflation in the 1970s the Iraq war spending is giving us the same thing today. I hate be a portrayer of doom and gloom, but I see no painless way out of our situation. Our best hope is drastic cuts in government spending, but that too will be painful.
markg
Nov 2 2007 at 11:43am
The Fed can create any bank reserves required for lending. The public’s desire to hold currency instead of demand deposits does not restrict bank lending.
Patrick R. Sullivan
Nov 2 2007 at 11:55am
That’s what happened during the Great Depression, as Anna J Schwartz just reminded Paul Krugman.
Bill Woolsey
Nov 5 2007 at 6:06pm
The quantity of money relative to the quantity of all liquid assets is irrelevant to the potential of bank runs to lead to monetary desruption.
What happens when there is a shortage of currency? What happens with there is a shortage of currency and checkable deposits.
The effort of each indidual to obtain these things involves reduced spending and selling assets. And generalized effort to sell (and little interest in buying) will reduce the liquidity of other assets. Further, collecting on debts as they come do generates money too.
Explain exactly where alternative sources of liquidity will appear to offet these effects? They don’t.
All I can figure is that you are confusing an analysis of what happens when there are bank runs with a critic of using bank reserves to predict total bank lending or maybe control total bank lending.
Comments are closed.