Macroeconomists think of the zero lower bound liquidity trap as something that only became a problem in 2008 (Japan aside). JKH, a commenter on this blog, says it became a problem for banks about 15 years ago.
"An interesting aspect of gap management is that the zero bound for interest rates has been a critical issue for banks for the past 15 years. This is because historic spreads between the prime rate and rates on some of the more passive core deposit and transaction accounts were quite wide. Spread compression began long before the fed funds rate or the Canadian bank rate got down to current levels. The zero bound isn’t a problem just for central banks."
Banking is a spread game. Banks borrow at lower interest rates, lend at higher interest rates, and make their income on the spread. Falling interest rates squeeze the bottom end of the spread against the zero lower bound, well before average interest rates approach zero. And the zero-bound spread-squeeze should affect financial markets more generally, not just financial intermediaries, in much the same way.
In macroeconomists' simplified view of the world, there are two assets: money and "bonds", where bonds includes everything else. (A slightly less simplified view has three assets: money, bonds, and real capital.) According to this simplified view, we only hit the liquidity trap when the interest rate on bonds hits zero. But if we view assets as a spectrum, with money at the bottom end, then falling nominal interest rates squeezes the whole concertina of interest rates against the lower bound, long before the top or middle of the concertina hits the lower bound.
Currency pays a zero nominal interest rate, and pays a real interest rate equal to minus the rate of inflation. Governments can finance their expenditures by issuing bonds at i% nominal interest, or by issuing currency at 0% nominal interest. Ignoring any costs of printing currency, or replacing worn and torn banknotes, we can say that i% represents the tax on holding currency: it is also the opportunity cost of holding currency rather than bonds.
Milton Friedman famously argued that the optimum quantity of money (or the optimum rate of inflation) would be found where the government eliminated the tax on currency by forcing inflation down low enough (to minus the real rate of interest) to make the nominal interest on bonds approach zero.
Some of banks' liabilities (chequable demand deposits especially) are close substitutes for currency. The closest substitutes for currency will pay interest rates closest to the zero interest rates that currency pays. They represent the bottom end of the banks' spread. One way to think of the spread-squeeze is to see central banks as lowering actual rates of inflation, and lowering expected rates of inflation, thereby lowering nominal interest rates on bonds, and lowering the tax on currency.
Central banks' product (currency) is in direct competition with commercial banks' product (chequable demand deposits). Both are media of exchange. When central banks lower the price of their product this harms commercial banks. (Falling nominal rates caused by falling real rates, due to rising supply of savings or falling demand for investment, could have had similar consequences.)
What are the consequences of this spread squeeze?
1. As the lowest of banks' interest rates approached the lower bound, a 1% reduction in the Bank of Canada's overnight rate, even if it were expected to be permanent, would not be fully reflected in banks' lending rates. It would be impossible for banks to lower interest rates by 1% across the whole spread.
"By the way, zero bound spread compression has something to do with the recent reluctance of Canadian banks to match bank rate decreases fully with prime rate decreases."
2. As the spread between borrowing and lending rates became squeezed, small changes in interest rates would have a bigger proportionate effect on the spread, and on banks' net income. Banks would have a bigger incentive to insure against this risk by using interest rate swaps.
"This put bank interest margins under threat of compression in a low rate environment – unless steps were taken. Some banks started hedging some of this interest rate exposure with interest rate swaps (on a sort of delta hedging probability basis). Interest rate swaps became exceedingly useful risk management tools around this time."
3. There are two ways that ultimate savers can lend funds to ultimate borrowers. The first is via a bank. The second is every other way (the shadow banking sector). One advantage banks have over shadow banks is that some of banks' liabilities are media of exchange. When central banks reduced the tax on currency, it reduced the profitability of the competing product offered by banks, and so reduced the size of the banking sector relative to the shadow banking sector. Since the total amount of borrowing and lending did not decrease, this meant the shadow banking sector grew.
4. As central banks reduced inflation and nominal interest rates, and reduced the variability of inflation, money became more attractive relative to other assets. Holding real, risky assets became less attractive, so the supply of loanable funds for real investment projects fell, at given real interest rates. To prevent a recession, central banks needed to reduce nominal and real interest rates still further, to prevent real investment and aggregate demand from falling below aggregate supply. As real interest rates on risky assets fell, the prices of real assets rose.
5. Weird things start to happen as nominal interest rates approach zero, even if we can keep the economy at a full-employment classical equilibrium. Money is like other assets, except (by definition) more liquid. (Money is usually safer as well, unless uncertainty over inflation is a concern, and that uncertainty has diminished anyway as central banks made inflation less variable, as well as lower.) So as nominal interest rates fall, the demand for money, relative to other assets, will increase without bound. If people save towards a target level of real wealth (as they do in overlapping-generations models), this means that in the limit, as nominal interest rates approach zero, people will want to hold all their wealth in the form of money, and none in other forms.
If by "money" in the above paragraph we mean money in the narrowest sense – "currency", then central banks must own all the real assets in a zero interest rate economy, since people always prefer money to real assets. If we widen the definition and by "money" in the above paragraph we mean demand deposits, then commercial banks will have to fund all their loans with demand deposits only, which creates serious problems for banks, since they must finance long loans using only the shortest of liabilities. If we widen "money" still further, to include highly liquid Treasury Bills, then governments must own all the real assets. Perhaps we are now seeing that theoretical curiosum in action.
Worthwhile Canadian Initiative 
Hmm. I wonder how this story connects with the puzzle of the 15-year asset price bubble?.
Must think on this.
This is much to assimilate.
“If by “money” in the above paragraph we mean money in the narrowest sense – “currency”, then central banks must own all the real assets in a zero interest rate economy, since people always prefer money to real assets”
Is this an argument against ZIRP? Would it be better to raise actually raise interest rates to something non-zero, admit the monetary policy ain’t going to work, and go full throttle with fiscal policy?
Even I understand spending gov’t money on sewers, schools, hospitals etc to stimulate the economy, but ZIRP is sounding like a crazy and scary place I don’t want to visited.
Patrick: It is certainly an argument against deliberately pushing inflation so low (i.e. negative) that you end up in a ZIRP (in other words, it’s an argument against Milton Friedman’s ‘Optimum Quantity of Money’, though whether Friedman intended it as a practical policy recommendation is another question, since it contradicted his other practical policy advice). But if you find yourself in a ZIRP by accident, where a short-run ZIRP is the only cure for a long-run ZIRP, then no.
Here’s the strange thing about interest rates: starting in equilibrium, with steady inflation, if the central bank cuts the rate of interest, inflation will begin to rise, and keep on rising, and the central bank will have to raise interest rates, not just to where they were before, but above where they were before, to keep inflation steady at a new higher level. Low interest rates cause high interest rates; high interest rates cause low interest rates.
Nick,
Recap: Central banks face the zero bound for policy interest rates, which are a type of wholesale deposit rate (i.e. interbank deposit rate). Commercial banks face the zero bound for rates on retail deposits of a particular type – those that are a near substitute for currency. These are sometimes called core deposits. Because core deposit rates are typically lower than the central bank policy rate, they hit the zero bound before the policy rate does. So with disinflation and declining interest rates, each of policy rates, retail rates, and retail spreads head toward the zero bound.
There is no comparable zero bound compression risk for the spread between wholesale rates and higher lending rates. Zero bound deposit spread compression may end up being subsidized by unbounded lending spread expansion (e.g. prime rate not following the bank rate down 1:1).
Krugman just wrote an interesting post on the zero bound, the yield curve, and interest rate optionality:
“But here’s the thing: the Fed can’t cut rates from here, because they’re already zero. It can, however, raise rates. So the long-term rate has to be above the short-term rate, because under current conditions it’s like an option price: short rates might move up, but they can’t go down.”
http://krugman.blogs.nytimes.com/2008/12/27/the-yield-curve-wonkish/
Core deposit pricing at the zero bound includes a similar optionality. Core deposits are usually the focus of “maturity transformation” discussions, where bank liquidity risk and deposit runs are the issues. But the zero bound for core deposit interest rates means there’s a different type of problem in interpreting their interest rate risk. So long as the deposit base is stable, and the zero bound is binding, they behave like long term fixed rate funding. But there is still the (option) risk that their interest rates might go up in the future. So the history of declining interest rates means that bank liabilities have tended to become longer in their effective duration (based on pricing rather than the risk of bank runs). Insurance companies have faced a similar problem, but more serious, in hedging their long term actuarial liabilities.
I’m not so familiar with the monetary theory you refer to, but it seems to me that asset bubble deflations have driven much of the risk aversion that is leading people to desire money rather than risk assets. Risk free interest rates have collapsed while risk premiums and the cost of capital have exploded. We have zero interest treasury bills, record corporate bond spreads, and Canadian banks doing equity issues at very depressed stock price levels. Does Friedman’s theory take into account this sort of boom and bust dynamic?
As bank interest margins came under general pressure as described, banks diversified their business model into things like wealth management and investment banking and became less reliant in a relative sense on the core businesses of deposits and lending. At the same time, capital requirements and capital management came under much more scrutiny, including the Basel capital guidelines. Ironically, it was the focus on capital requirements that allowed banks to crank up the securitization machine and move at least some of their funding for mortgages and credit cards off balance sheet.
I find the strange thing about interest rates is that conventional monetary policy success by definition will drive nominal and real rates toward zero, where the balance between inflation and deflation is most precarious, and therefore where the risks for monetary policy stability become the greatest – sort of like an expiring option, where the swings in option “gamma” can be very wild and difficult to control.
Sorry if it’s somewhat OT for this thread ….
Excellent post by Brad Setser today:
http://blogs.cfr.org/setser/2008/12/29/the-collapse-of-financial-globalization/
From his post:
“Central banks were the main source of financing for the US deficit all along.*** Setting Japan aside, the big current account surplus countries were all building up their official reserves and sovereign funds — and they were the key vector providing financing to the deficit countries”
With China experiencing a ‘hard landing’, and the Gulf states reeling from low oil prices it looks to me like those who financed the deficit countries (esp. the US) are going to be less able and/or willing to do so in the future? China, Korea, etc. will have to spend their reserves to boost domestic demand. Similarly, the Gulf states will have to spend their war chests at home to finance program spending and stimulus at home. It seems unlikely that private money will fill the resulting hole. If this is the case, then how long will the Fed be able to maintain ZIRP? Or do they just print the difference?
JKH: Sorry for the delay; Christmas and all. I think we are on the same page. Friedman’s Optimum Quantity of Money paper is silent on the question od short-run fluctuations (as far as I can remember); it’s more about the long-run. It has been very influential in monetary theory, but I’m not sure how influential it has been for actual monetary policy, other than one argument among many for getting the average rate of inflation low. So far, the Bank of Canada has only been targeting 2% inflation, which is certainly lower than inflation in the 1970’s and 1980’s, but still higher than the minus 2% that Friedman’s paper would argue for.
Successful monetary policy has been defined as bring down inflation, which will bring nominal interest rates down, but should not have a first-order effect on real interest rates.
Patrick: off topic, but interesting nevertheless. I think the Fed is quite willing to support ZIRP by printing as much money as is needed, until the ZIRP is no longer needed.