This is exactly what we wanted to prevent
Wednesday, March 26, 2008
Yes, Virginia T-Bill rates can be negative.
In economic theory and practice we typically consider this an impossibility because agents would prefer to hold cash.
However, holding cash is not costless - particularly not now. Holding physical cash in the quantities needed by hedge funds and sovereign wealth funds has obvious costs. You have to build a vault, guard the vault, protect against employee theft, etc.
However, cash accounts at financial institutions also impose a cost if you believe that there is a liquidity crisis in the works. If a bank or investment house fails, then at best your account is likely to be frozen for some period of time. At worst you may suffer permanent losses.
If it turns out that you can't hold physical cash, you are nervous about large cash accounts and similar liquid markers such as Auction Rate Securities are seizing up, then it might be worth it to pay the US government hold your cash.
Now, I don't think that T-Bill rates will go negative because I think that banks, who can realistically hold high quantities physical cash will sell them aggressively. Yet, it is important note (in answer to Krugman's query) that if bank holdings are the channel by which the T-Bill rate stays above zero then monetary policy is not impotent in a zero rate T-Bill world.
Posted by Karl Smith at 1:01 PM
Monday, March 24, 2008
Rumor is JP Morgan is thinking about upping the ante on the Bear Stearns deal to help it go through. The Fed should discourage this.
The original sale price of Bear was $2 a share and represented a virtual wipe out of Bear shareholders. Given the extensive role the Fed played in staving off bankruptcy this was appropriate. The Fed should in no way signal that it is prepared to step in to save stockholders. It should step in only to assure smooth functioning of credit markets.
However, now JP Morgan seems to be caving from pressure from Bear shareholders who may vote down the deal out of spite. The Fed cannot stop this directly but it can issue statements which imply its approval of the $2 price.
I would say something like:
In other words, Bear shareholders are lucky to have $2. Please do not make things worse than they already are.
The funding and guarantees offered by the Fed were supported by JP Morgan's unique ability to provide stable assurances to Bear Stearn's clients, creditors and counterparties. We anticipate that few other situations could have or will justify such measures.
Posted by Karl Smith at 9:48 AM
Friday, March 21, 2008
What options do we have for stimulating the economy once T-Bills rates hit zero? Traditionally, the Fed encourages bank lending by buying T-Bills.
The Federal government limints the amount of loans a bank can give out based on the amount of cash they have in the vault. So each day the bank has to decide to
A) Keep money in the vault and use it to support new loans to customers
B) Loan out the vault cash to other banks
C) Loan the vault cash to the Federal government by buying government bonds also known as T-Bills
To stimulate the economy the Federal reserve buys up T-Bills. This lowers the interest rate on T-Bills and encourages banks to try to lend more to each other. However, on net inter-bank lending must be zero. That is, every loan to one bank must be matched by borrowing by another bank.
When too many banks try to lend to each other at once the interest rate on interbank lending, the federal funds rate, falls. This leaves the banks with only one good option - lend more to customers. And that exactly what we want.
What happens, however, if the interest rate on T-Bills goes to zero. Then even if the Federal Reserve buys more T-Bills it doesn't encourage banks to lend more to each other. Thats because the Fed is not making T-Bills less attractive. Zero is still zero.
One question might be, "If the interest rate on T-Bills is zero but the Fed funds rate is not zero, then why would ever buy T-Bills."
The answer is that T-Bills are backed by the full faith and credit of the US government. When a bank loans to another bank it has the full faith and credit of that bank, but in this environment the banks don't have the same kind of credit they used to.
So what can the Fed do to get around this?
It can lend to the banks directly. Currently the Fed charges more to lend money to banks than banks charge each other. Thats because the Fed only wants lend as a last resort.
However, the Fed could change that and make borrowing from the Fed just as attractive as borrowing from other banks.
Some economists worry that banks will be hesitant to borrow from the Fed because typcially that is seen as a last resort and no bank wants to look desperate.
However, as the cost of borrowing from the Fed goes to zero and possibly even (gasp) below zero banks will change their minds. Indeed, they will probably all change their minds at once in what researchers think of as a tipping point.
The question is how do we manage that tipping point. One way is to make the loans a temporary feature. The Fed can allow temporary loans for up to 28 days and less than say $50 billion at this super low rate. Then it can judge participation and if necessary offer new loans when the 28 days is up.
If these rates are low enough banks will find them irresistible. Its possible that for short periods the Fed could even loan out money at negative rates. That means the Fed pays banks for taking out a loan. Thats an offer that few banks could refuse.
Posted by Karl Smith at 11:18 AM
Wednesday, March 19, 2008
KNZN mentioned that he didn't think real yields could collapse enough to produce a liquidity trap in an inflationary environment.
The 90 day T-Bill is now at half of a percent, and looks to be in freefall. Remember this is after the Fed annouced that I-Banks would have access to the discount window and that Bear Stearns was brougt back from the brink of bankruptcy.
Had Bear gone under could interest rates have remained above zero?
There is no tightrope. We should starting talking about making the financial sector an offer it cannot refuse. More on that tomorrow.
Posted by Karl Smith at 1:30 PM
Tuesday, March 18, 2008
I am off to the state legislature this morning so no faux statement. However, I still maintain that there is no tightrope. The Fed has to be focused on preventing the liqudity trap and jump starting credit markets as soon as possible.
Moreover, M1 is flat and credit contracting. This should be leading to an effective decline in the money supply. Ultimately that is deflationary.
The statement should have some nod to commodity prices and risks but the predominate concern is stability in financial markets and the outlook for growth. In short, look out below, we are heading for 1% as fast as is prudent.
Posted by Karl Smith at 10:23 AM
Monday, March 17, 2008
I am a big fan of what appears to have gone down in the Fed - JPMorgan - Bear Sterns deal. It worked out just about as well as it could given the situation.
Bear Stearns shareholders lost everything, which helps prevents moral hazard. Creditors, counterparties and clients were saved which keeps the financial system functioning.
When the dust settles this could be one central bankings finest moments.
Posted by Karl Smith at 12:15 PM
Saturday, March 15, 2008
You really have got to hand it to the financial press. After stoking the fires of risk fueled housing and equity bull markets they have now fully transitioned into fanning the flames of widespread panic.
Its a wonder that more banks don't tap the discount window after the confidence inspiring conniption that virtually every financial pundit seemed to go through yesterday.
Was there a major crisis at Bear, of course. By no means am I suggesting anything less. However, the term "FED Bailout!!!" was splashed across every TV and computer screen that I saw, when the reality was somewhat more prosaic.
About three-quarters of the way down this Wall Street Journal article the author admits that
technically the Fed still hasn't lent directly to investment
the New York Federal Reserve Bank had agreed that it would provide financing to Bear Stearns via J.P. Morgan Chase. J.P. Morgan Chase was used as a conduit because, as a commercial bank, it already has access to the Fed's discount window
So basically, as I see it, the Fed encouraged JP Morgan to use the discount window essentially as intended - to stem a bank run.
Now, it wasn't a bank that was being run but in our modern financial system the Fed understands that intermediation extends far beyond traditional banking institutions. Yet, the Fed still went through a member bank to conduct this operation so as to avoid mudding the waters as much as possible.
As far as I can tell the big innovation was that Fed guaranteed Morgan against losses, which given the time frame available seems not only prudent but reasonable.
Posted by Karl Smith at 9:23 AM
Friday, March 14, 2008
Lots of people have said to me both on and off line that we don't have to worry about the Japan Scenario because we have a solid inflation buffer in the US.
While the inflation buffer gives us more room in a sense, it is important to remember that it is not deflation per se that causes a liquidity trap. It is that the equilibrium interest rate is below zero.
It is possible that the equilibrium risk free interest rate is a real negative 3% in this crisis, which implies that we still won't be able to get there with 2.7% inflation.
Exploding risk premiums could drive the equilibrium real rate that low because what matters is credit availability to firms and consumers.
So we are not in a position were we can ignore the liquidity trap possibility. On top of that is the issue that there are increasing deflation pressures in the decline collateral values, falling consumption and the potential for dramatically slower global growth. While ultimately they might not override inflationary effects of recent Fed policy, they are not to be ignored.
In short deflation cannot be ruled out and the liquidity trap remains a threat even in a moderately inflationary environment.
Posted by Karl Smith at 9:33 AM
Some of my fellow bloggers have been looking at inflation expectations and are worried that the Fed is pumping too much cash.
I am skeptical.
Inflation ultimately has to come through the interaction of supply and demand. Money creation may be the source of surging demand and hence higher prices but demand has to in fact surge. It is difficult to have Fed driven inflation in an environment of falling retail sales.
If consumers contract then businesses loose pricing power. Commodities ultimately have to follow business demand.
Posted by Karl Smith at 8:06 AM
Monday, March 10, 2008
I may be one of the few bloggers who think the Feds policy actions are now moving in exactly the right direction. My long standing view is that we are experiencing the popping of a massive credit bubble that extends far beyond subprime mortgages.
A credit bubble happens when the price of credit becomes to low rather than too high. Low priced credit like asset bubbles can be self-reinforcing. Cheap credit means that people can easily refinance their old debts. This puts more money back into the market at lower risk levels which further encourages cheap credit.
See, essentially the price of credit is determined by two things. First, is the scarcity of money. When you loan someone money, you can't use it to buy things yourself. Therefore, few people want to loan out all of the money they have. However, if the government prints more money, more can be loaned out without running short.
Controlling the quantity and hence the scarcity of money is usually what the Fed does to control the price of credit.
However, there is another factor, default risk. Even if you didn't need your money right now you might be hesitant to loan it out if you thought you wouldn't get it back. Hence, the price of credit rises when people think they might not be repaid. The difference between the price of risk free loans and risky loans is the credit spread.
So, how does this affect the current situation. The subprime explosion and subsequent implosion happened because of collapsing credit spreads. That is, investors stopped worrying as much about getting their money back. In particular they believed they had fancy computer simulations which assured that they could make money even if lots of the loans defaulted. As long many of them did not, it wouldn't be a problem.
Now it turns out that those simulations were based on a world in which credit was relatively hard to come by. This new world, in which investors were loaning to anyone with a pulse operated by different rules. In short, the computer simulations lead to the creation of a world in which the simulation was no longer valid.
Computers in some ways are like extreme autistics who often suffer from these types of recursive mistakes. They fail to realize that the response pattern of the outside world is dependent on the outside world's perception of the computer's response pattern. Thus an infinite cycle of you think, that I think, that you think, . . . is set up. Cognitively normal people can solve these problems intuitively but they form a serious calculation issue for computers and autistics. But, I digress.
The important thing is that the simulations failed to account for the changing nature of the credit market. Thus, investors took on way more risk than their analysis had predicted. In late 2006 and early 2007 it started to become clear that the analysis was wrong. By the summer of 2007 credit spreads were spiking as investors tried to undo what they had done.
The Fed, at first tried to fix the problem by increasing the supply of money. The hope was that this could drive down the price of credit, allow people to refinance and investors to get out of the bad deals they were in. However, the deals were going bad too fast. The Fed could not keep up and despite their best efforts spreads are growing again.
So what do?
The Fed must attack the credit spreads not just the scarcity of money. This is what it is doing now by taking an increasing variety of assets as collateral for up to $200 Billion in loans. What that means is that the Fed is allowing banks to borrow cold hard cash by using your risky asset as collateral. In addition, its going to keep rolling this loan over indefinitely.
This means that some of the risk has shifted from the bank to the Fed. If the asset goes bad and the bank can't repay its loan to the Fed, the Fed ends up with the bad asset. In a sense it implies that risky assets are only so risky because at the end of the day you can always pawn them off on the Fed.
Doesn't this put the Fed at risk? Well, the Fed buys the asset with money that it creates out of thin air. Things get a little complicated when one considers that the Fed has to also keep an eye on the Federal Funds rate, which it manages by creating money out of thin air. Yet, I don't think thats much of a problem. This post is already longer than I intended so why its not a problem will have to be saved for another day.
Posted by Karl Smith at 10:29 AM