Toro
Diamond Member
I'm going to try to answer at least some of your questions in the posts that follow. They are, of course, very good questions, ones that require a fair amount of thinking. At the end of the day, I am often mentally exhausted, and come here as much not to think and put my mind on autopilot as anything. So I will probably give short shift to answers that deserve more.
I think there are a number of things.
First, my general belief is that most macroeconomic models are fine for describing an economy under "normal" conditions but break down during abnormal conditions because the assumptions about rationality do not hold. For example, you say that policy is "tight." That may be the case under a Taylor rule, but that assumes that human behavior is the same below the zero bound as it is above the zero bound under normal conditions. I'd argue differently. I think human behavior changes. The effects of interest rates at near zero alter the behavior of savers, forcing them out on the risk curve in ways that they would otherwise not undertake, let alone a -7% theoretical rate that may be necessary in a theoretical Taylor rule.
Most standard macroeconomic also assume that asset prices do not have an affect on the broader economy. I think this is an extremely critically flawed assumption, and is an utter failure of economics as a discipline. The real Fed Funds rate has averaged less than zero since 2000, which to me is a damning indictment of the failure of traditional macroeconomics. I think Minsky and Kindelberger more accurately describe the role of asset markets in the economy than, say, a Cochrane or a Roll.
In traditional macroeconomics, assets and liabilities net out on both sides of the equation, so theoretically should have no affect on the economy. Turns out, this isn't the case, even from traditional macroeconomics. There is a Fed paper somewhere, I believe, which concluded that for every $1 increase in the stock market wealth, it increased consumer spending by 7 cents. Again, that may be true under normal conditions, but when the tails get fat in ways the Gaussian assumptions of the regression model variables do not account for, the effect is more profound, particularly on the downside.
We do know that there is at least some truth to that. Normal distributions around the mean assume symmetry but human behavior does not comply. We know empirically that the pain of loss is 2-3x greater than that of gain. And my anecdotal guess is that this dispersion is even greater when there are tremendous shocks of loss, such as we saw in the Fall of 2008.
The answer to your question is because supply does matter. It is not just a matter of demand. Supply has to get wiped out before the economy can get back to long-term trend. This, I believe, is rooted in neo-classical theory articulated by Alfred Marshall, who noted that the long-term equilibrium rate of interest should equal the accumulation of the capital stock, since the cost of money, i.e. interest, is the opportunity cost of investment. I believe this is true, over the long-run. It is not necessarily true in the short run, however because the behavior of human beings does not always follow the assumptions in macroeconomics - that human beings are hyper-rational beings able to calculate probability distributions across infinite utility curves.
Kindelberger believed that asset bubbles occurred when there are exogenous shocks to the system. I believe there were two enormous exogenous shocks over the past decade - the commercialization of the Internet and the opening of China. What happens when you have a shock is that there is an abrupt change. Let's look at the Internet bubble. In this case, the abrupt change lowered cost curves as traditional middlemen and wholesalers were and still are being disintermediated. For the first Net entrepreneurs, their profit margins were enormous, and they became extremely wealthy very fast. In traditional theory, what should happen is that those profit margins should be wiped out as more entrepreneurs enter markets. And over the long-run, this is true. But it does not happen fast enough as the models suggest it should.
In doctrinaire macroeconomic theory, there are no positive feedback loops. There are only negative feedback loops. For example, there can never be excess profits because arbitragers or entrepreneurs will wipe excess profits out. Again, that is true in the long run but not in the short run. In the short run, and even the intermediate term, there are positive feedback loops, sometimes powerful ones. So excess profit margins create irrational behavior. And the greater the rewards, the greater the frenzy. So 28 year-old billionaires creating fortunes out of products that did not exist five years ago create manias that traditional macroeconomics say cannot exist.
Tying it back to Marshall's theory, the accumulated capital stock starts growing faster than the long-term trend because entrepreneurs view their profit margins as permanent, which leads to over-investment. I saw this en masse when I ran my fund. I used to run a $500 million equity portfolio during the 90s, and saw countless analysts modeling all these brand new companies that didn't exist a decade prior (and don't exist today) making unGodly amounts of money indefinitely. And the capital markets threw an ocean of money believing them. Capital markets acted in ways that dogmatic market proponents said wouldn't happen. They lost their mind. I can give you story after story about the insanity during that time. And we poured hundreds of billions, if not trillions of dollars into companies that were no more than illusions.
The Federal Reserve should have seen this, but instead Greenspan cheered it on. The Fed should have been raising rates to make capital more dear, not cutting rates as it did in 1998 during the LTCM melt-down. Now, I do have some sympathy with Greenspan's argument that raising interest rates to pop a bubble can have worse side affects than the bubble, but what should have happened under Kindelberger's theory is that as the perceived profit margins rose, the Fed should raised interest rates to whatever the perceived rate of the accumulation of the capital stock was for the market to stop the frenzy. If entrepreneurs had known that the cost of capital was higher and their profit margins lower, we would have had less accumulation of the capital stock.
So what happened was there was a frenzy of building, particularly in unlit fiber that carried Internet traffic. Over the long-run, this wound up not being a problem - eventually markets get it right - since the Internet visionaries and dreamers wound up being right. That fiber is now being lit, but at the time, it plunged the technology industry into a true depression.
What solved the problem was time and absorption of supply. I get that when demand collapses, the government should step in to absorb it, at least to some extent. But the underlying structural problem is not demand. It is supply. The long-run growth of the capital stock should equal to the long-run growth in the economy. If the growth in the capital stock exceeds that of the economy for some period of time, eventually, supply growth will collapse and the economy has to recalibrate and absorb the excess capacity. Essentially, all economic growth equals productivity growth plus population growth. So as long as both are growing, eventually we will absorb the excess capacity, but lowering interest rates to -7% isn't going to absorb the excess capacity because the excess capacity must eventually come back in line with long-term economic growth, which means a period of time of low aggregate economic growth as the supply gets absorbed.
Anyways, that is a long and meandering answer to your question.
Why do you think an asset bubble collapse would have such a pronounced effect on the broader economy? The crucial thing about this bubble was that everybody became highly leveraged to buy into it. As you point out, there's too much debt overhang and everybody is still trying to deleverage. But why does deleveraging effect the real economy? If everybody is trying to save more, and on aggregate we have an excess of savings and a shortage of investment, why doesn't the interest rate just fall to equilibrate savings and investment, restoring full employment? The zero lower bound comes in here. Does that mean monetary policy can't do anything since nominal interest rates can't drop further? Of course not. They can still lower the real interest rate as much as they like by creating expectations of a higher price level/NGDP level. Bottom line, the bubble story effects the economy through aggregate demand, not its supply potential, and aggregate demand can always be controlled by monetary policy.
A more intuitive way to think about deleveraging is that NGDP, or nominal income, is the resources the economy has to pay down nominal debts. But the Fed allowed NGDP to fall 9% below trend in 2008! Our ability to deleverage has been sabotaged by a tight monetary policy.
I think there are a number of things.
First, my general belief is that most macroeconomic models are fine for describing an economy under "normal" conditions but break down during abnormal conditions because the assumptions about rationality do not hold. For example, you say that policy is "tight." That may be the case under a Taylor rule, but that assumes that human behavior is the same below the zero bound as it is above the zero bound under normal conditions. I'd argue differently. I think human behavior changes. The effects of interest rates at near zero alter the behavior of savers, forcing them out on the risk curve in ways that they would otherwise not undertake, let alone a -7% theoretical rate that may be necessary in a theoretical Taylor rule.
Most standard macroeconomic also assume that asset prices do not have an affect on the broader economy. I think this is an extremely critically flawed assumption, and is an utter failure of economics as a discipline. The real Fed Funds rate has averaged less than zero since 2000, which to me is a damning indictment of the failure of traditional macroeconomics. I think Minsky and Kindelberger more accurately describe the role of asset markets in the economy than, say, a Cochrane or a Roll.
In traditional macroeconomics, assets and liabilities net out on both sides of the equation, so theoretically should have no affect on the economy. Turns out, this isn't the case, even from traditional macroeconomics. There is a Fed paper somewhere, I believe, which concluded that for every $1 increase in the stock market wealth, it increased consumer spending by 7 cents. Again, that may be true under normal conditions, but when the tails get fat in ways the Gaussian assumptions of the regression model variables do not account for, the effect is more profound, particularly on the downside.
We do know that there is at least some truth to that. Normal distributions around the mean assume symmetry but human behavior does not comply. We know empirically that the pain of loss is 2-3x greater than that of gain. And my anecdotal guess is that this dispersion is even greater when there are tremendous shocks of loss, such as we saw in the Fall of 2008.
The answer to your question is because supply does matter. It is not just a matter of demand. Supply has to get wiped out before the economy can get back to long-term trend. This, I believe, is rooted in neo-classical theory articulated by Alfred Marshall, who noted that the long-term equilibrium rate of interest should equal the accumulation of the capital stock, since the cost of money, i.e. interest, is the opportunity cost of investment. I believe this is true, over the long-run. It is not necessarily true in the short run, however because the behavior of human beings does not always follow the assumptions in macroeconomics - that human beings are hyper-rational beings able to calculate probability distributions across infinite utility curves.
Kindelberger believed that asset bubbles occurred when there are exogenous shocks to the system. I believe there were two enormous exogenous shocks over the past decade - the commercialization of the Internet and the opening of China. What happens when you have a shock is that there is an abrupt change. Let's look at the Internet bubble. In this case, the abrupt change lowered cost curves as traditional middlemen and wholesalers were and still are being disintermediated. For the first Net entrepreneurs, their profit margins were enormous, and they became extremely wealthy very fast. In traditional theory, what should happen is that those profit margins should be wiped out as more entrepreneurs enter markets. And over the long-run, this is true. But it does not happen fast enough as the models suggest it should.
In doctrinaire macroeconomic theory, there are no positive feedback loops. There are only negative feedback loops. For example, there can never be excess profits because arbitragers or entrepreneurs will wipe excess profits out. Again, that is true in the long run but not in the short run. In the short run, and even the intermediate term, there are positive feedback loops, sometimes powerful ones. So excess profit margins create irrational behavior. And the greater the rewards, the greater the frenzy. So 28 year-old billionaires creating fortunes out of products that did not exist five years ago create manias that traditional macroeconomics say cannot exist.
Tying it back to Marshall's theory, the accumulated capital stock starts growing faster than the long-term trend because entrepreneurs view their profit margins as permanent, which leads to over-investment. I saw this en masse when I ran my fund. I used to run a $500 million equity portfolio during the 90s, and saw countless analysts modeling all these brand new companies that didn't exist a decade prior (and don't exist today) making unGodly amounts of money indefinitely. And the capital markets threw an ocean of money believing them. Capital markets acted in ways that dogmatic market proponents said wouldn't happen. They lost their mind. I can give you story after story about the insanity during that time. And we poured hundreds of billions, if not trillions of dollars into companies that were no more than illusions.
The Federal Reserve should have seen this, but instead Greenspan cheered it on. The Fed should have been raising rates to make capital more dear, not cutting rates as it did in 1998 during the LTCM melt-down. Now, I do have some sympathy with Greenspan's argument that raising interest rates to pop a bubble can have worse side affects than the bubble, but what should have happened under Kindelberger's theory is that as the perceived profit margins rose, the Fed should raised interest rates to whatever the perceived rate of the accumulation of the capital stock was for the market to stop the frenzy. If entrepreneurs had known that the cost of capital was higher and their profit margins lower, we would have had less accumulation of the capital stock.
So what happened was there was a frenzy of building, particularly in unlit fiber that carried Internet traffic. Over the long-run, this wound up not being a problem - eventually markets get it right - since the Internet visionaries and dreamers wound up being right. That fiber is now being lit, but at the time, it plunged the technology industry into a true depression.
What solved the problem was time and absorption of supply. I get that when demand collapses, the government should step in to absorb it, at least to some extent. But the underlying structural problem is not demand. It is supply. The long-run growth of the capital stock should equal to the long-run growth in the economy. If the growth in the capital stock exceeds that of the economy for some period of time, eventually, supply growth will collapse and the economy has to recalibrate and absorb the excess capacity. Essentially, all economic growth equals productivity growth plus population growth. So as long as both are growing, eventually we will absorb the excess capacity, but lowering interest rates to -7% isn't going to absorb the excess capacity because the excess capacity must eventually come back in line with long-term economic growth, which means a period of time of low aggregate economic growth as the supply gets absorbed.
Anyways, that is a long and meandering answer to your question.
Last edited: