Immediately preceding the worst of the financial crisis, my professors all pointed out to me that the yield curve had inverted -- short-term yields were more risky than 20-year or 30-year Treasury securities. My professors all seemed nervous about this.
Aside from such a situation being generally counter-intuitive, why was this viewed so negatively? What kind of effects does this have, in a more macro sense?
Nano answer: declining yield curve is a symptom that the market expects the economy to worsen.
Short answer: yield curve is built on expected future interest rates. Yield curve will be downward sloping if you expect interest rates to fall in the future. Interest rates fall because the central bank react to worsening economy.
So if the yield curve is downward sloping it means the market expects the economy to worsen to the point that the central bank will have to intervene.
To explain why a negative sloping yield curve is bad, you have to start with a theory of the yield curve.
The dominant theories for the term structure of interest rates are the rational expectations, liquidity preference, and market segmentation. (The first two theories are quite compatible with each other and have more standing so let's assume that view.)
Generally, we expect longer-term maturities to have correspondingly higher yields because the investor must bear more risk (inflation risk, bond price risk, interest rate risk, liquidity risk, credit risk). So it is indeed "abnormal" to observe an inverted yield curve -- why would an investor hold longer-term maturities in the face of these risks? The answer is that that long-term investors expect the term-structure of rates to drop even further. These expectations can be measured in the implied forward rates of the term structure.
What is the negative slope mean for economic growth? The negatively sloped yield curve is a function of the short-rate and longer-rates. The Federal Reserve targets the short-rate thru open market operations (purchases and sales of short-term treasury bills). The Fed may raise the short-rate substantially to limit credit growth.
In this world, bank have less incentive to lend long and fund capital or residential investment projects since banks can lend short and capture a higher rate. This reduces the pace of credit growth and the number of new projects financed by debt. Also, the risk-free rate is an asset class that competes with equities and other asset classes (housing, gold, etc.). The higher the short-term rate the less speculation in other asset classes which can lead to bubbles bursting.
Note that banks will still continue to issue loans at longer maturities, however, they will focus on higher quality investment-grade credits. Firms that have marginal access to credit may experience re-financing difficulties and may not benefit from lower long-term risk free rates since credit spreads typically widen in a recession.
Also, there is research showing that long-term treasury rate is a proxy for nominal economic growth. If the long-rates are very low and the short-rate is high, the bond market is anticipating weak economic growth and future Fed rate cuts.
Theoretically, a rising yield curve is compensation for the additional duration risk.
An inverted yield-curve is saying that the market thinks that:
Next-year's figures for: growth plus inflation
is less than
Ten years' time's figures for: growth plus inflation
Which means that expectations are either of a recession (some negative economic growth; and as far as the indicators are concerned, at least two consecutive quarters of negative growth), or that inflation in 10 years time is expected to drop relative to what's in the rates now.
The US had yield-curve inversions for less than three months in 1995 and 1998, without immediate subseequent recession, so it can be a false signal. The US yield curve inverted, when, mid 2005? So wasn't a reliable signal of "when". The yield curve inverted in 1966, and was not followed by recession. But the three recessions before that were not preceded by a yield-curve inversion; the 1990/1 recession wasn't either.
The UK yield curve spent a few years flat / inverted, in the noughties, during economic boom times (driven by high demand from pension funds for long-dated gilts).
So an inverted yield curve can be an unreliable indicator of recessions.
Samuelson once quipped that the yield curve had successfully predicted 9 of the last 5 recessions.
Explanations for why the yield curve inverts have been covered adequately above. But what does it mean for the economy? For the yield curve inversion to predict stock market performance enough to make an actual decision, the bond market would have to be more efficient than the stock market (ignoring carry costs, etc.). That seems like a pretty strong assumption.
An inverted yield curve basically means that interest rates will be higher for the coming year than for the years following.
That means that entities that need do borrow for short term purposes will do so at a greater cost that those borrowing for the long term. That is an unusual and "unnatural" relationship. All other things being equal, that will dampen economic activity in the coming year, likely producing a recession.
An interest rate "curve" functions something like a weather report. In the case of a inverted yield curve, it is forecasting "stormy weather just ahead."
The vast majority of credit is produced by banks.
Banks "borrow short, lend long", meaning: they borrow in the short term, mostly in the form of deposits that have a theoretical term of 0 or near 0 which pay the same, and they lend in the long term, mostly in the form of government debt, mortgages, and corporate bonds & loans for many years. The Fed produces this data for free. Banks have been fond of getting themselves into trouble with commercial paper through special purpose entities every other expansion or so, but most funding comes from deposits.
One might say "but banks only hold a fraction of total credit", and one'd be right, but that's not what the modern bank does. Before tight rule of law, agency law, and ease of financial trading, banks did hold most of what little credit existed in the world. Now, they are merely the producers of credit, actually only receiving ~50% of net income from net interest; the other ~50% comes from fees & selling those produced loans to investors at a profit.
After a banknote issuer (central bank, monetary authority, etc) has failed to maintain a level of inflation that it prefers to a large degree if it even cares at all, it usually overshoots in the other direction, selling assets/raising the price of whatever it prefers to control instead of controlling inflation directly, causing interest rates to rise with diminishing effect across the rest of the yield curve. This "other price" in the US, usually called the Fed funds rate, is the rate (spread) that banks charge each other for borrowing (trading) required reserves (deposits) at the Fed. Whatever price a banknote issuer prefers to control will always be controlled through one method: controlling the quantity of assets on its balance sheet.
If a bank has now lost ~50% and more of its net income because the net interest margin is less than the cost to produce credit, it shouldn't be expected to continue to produce credit. If credit is not being produced, investment cannot be funded. If investment cannot be funded, operations cannot be funded. If operations cannot be funded, wages & salaries cannot be funded...
