Note to viewers: This is a video replay of GMO chief investment strategist Jeremy Grantham's keynote address from the 2015 Morningstar Investment Conference in Chicago. GMO was not able to provide the presentation slides as a downloadable PDF, but some the slides are viewable in the video recording.
Jeremy Grantham: Hello, nice to be here.
Most of us spend a lot of our time working on unimportant matters--pretty boring. I know I did. But the good news is I don't anymore. I have 600 colleagues to do the heavy lifting, and I am totally free to obsess about important issues that interest me. This is a great job description.
I'm going to tell you what the 10 are, and these are questions you can ask me in question time. Anything else is out of bounds.
First of all, limitations of Homo sapiens. I am not pessimistic, by the way. It's you guys who are optimistic. What they found out is that, Homo sapiens have a strong predisposition to be optimistic, bullish and have a wicked capability of avoiding unpleasant information, which results in bubble after bubble without any memory of the pain that preceded it. It allows for ignorance of all the problems that come with resource limitations, and it allows for the denial of the facts on climate change, for example.
Two, longer-term resource limitations. Economists think the economy works with capital, labor, and a perpetual motion machine. They simply don't do limitation of resources. We run out of nothing in mainstream economics.
Three, especially oil. It's worth a separate point. Oil has driven our civilization to where we are. It allowed for all the scientific research and so on, all the surplus in our system. And now we're running out of cheap oil. It's a stress to the economy.
Four, climate problems. Breaking around our ears as we speak, California is the driest it's been in 1,200 years, according to tree rings. Houston, after six years of drought where farmers couldn't plant a single cotton crop, then gets seven years of rain in one go with unfortunate fatal flooding. Things are moving rather fast, papal and cyclical. The Pope, however he arranges it, is advised by a couple of dozen Nobel Prize winners, including the very best scientific advice on climate change that money can buy. Angela Merkel, hereafter known as "the Angel," arm-twisted the G7 into agreeing to a statement that they will be off fossil fuels by 2100. This is something that your clients will be increasingly interested in going forward into the next decade or two.
Five, food problems. Probably, the biggest problems that we face: water, soil erosion, population growth, climate extremes, flooding and droughts, increasing resource cost. This will destabilize, unfortunately, an increasing number of poor countries in my opinion.
Six, income inequality. It's very hard to imagine that an army of average workers making no progress is good for economic growth.
Seven, pressure on GDP growth in the developed world, which has been a big thing that I've been hacking on in recent quarters. Aging and slowing population has reduced the population effect in the U.S. by 1% since I arrived in the country. Resource constrains, low capex, which will come back, inequality itself, climate pressures, and the low-hanging fruit is mostly gone. Facebook is not the steam engine.
Eight, I am trying to understand the deficiencies in modern capitalism, democracy and corp-bureaucracy. The loss of corporate stakeholders. Once upon a time, corporations looked after their workers, looked after the city they were in, looked after the state, and the country, and now it is profit maximization, and to hell with other stakeholders. There is no ability in capitalism to treat very long-term tragedies of the commons--air, water, and soil. The discount rate guarantees that your grandchildren have no value, public opinion has no effect on the probability of a bill passing, according to professors at Princeton. The financial elite, on the other hand, when they hate a bill, it has a 5% chance of passing. When they love it, 60%. And we the people, whether we love or hate it, it has a 30% chance. This is not really effective democracy in action. And then the stock option culture, which I'll cover in more detail, where we buy stock back rather than build a new plant.
Nine, deficiencies in the Fed. They have a bad job description, badly executed, and have been a cause of constant trouble in the financial markets. They have created a steady stream of bull markets that end badly, and they brag about the wealth effect they get from that. All three of them of the Greenspan era--Bernanke and Yellen--they have all bragged about their ability to push up stock prices. In other words, they are admitting that the stock market is manipulated.
Finally, investment bubbles, which is what I am going to talk about today. It's always had a great fascination for me, and the question today is, are we in one? Is it a significant concept, and when is going to happen?
To set the scene, we're going to take a look at how the market works. The market is driven by career risk. Everyone's job description in the end is to keep your job. And in the stock market, you do it by following Keynes' advice: never be wrong on your own. If you're going to be wrong, make sure you've got plenty of company. And you do this by looking around you to see what other people are doing, and you do more or less the same thing. Keynes said the way to succeed is do what everyone else is doing, but beat them to the draw. Be just a little quicker and slicker than they are. This process guarantees that everyone herds together and creates enormous momentum, which drives every asset class way beyond fair value. In fact, it is overwhelmingly the major source of market inefficiency.
But the good news is that we live in a real world where there is arbitrage, and if you drive the stock price far away from replacement cost, you create a slow but inevitable arbitrage. So if you can sell in the stock market a new plant at three times the cost of building it, you'll build another two and sell them in the market in a process that will end with a huge amount of, let us say, fiber-optic cable, which is what happened in 2000.
If on the other hand, your factory is worth half of what you paid to build it, you will not build anymore; no one will build. There will become a shortage. Eventually, the prices will rise. One day you sharpen your pencil, and you'll find that you can finally build a new polyethylene plant with a safety margin, and you build it. So everything converges in the capitalist process on replacement cost. And that is the process that creates opportunities for value and makes regression work. So momentum pushes it away, and then it regresses back under the slow burn of capitalist process.
The big problem with this is timing. Sometimes an economic cycle in the chemical industry will occur in a year; sometimes it will take six or seven years. The client's patience--as you know better than me, or as well as me--is 3.00 years. So if you have a long, slow-burning cycle you can arrive at the winning line with a completely different set of clients or none at all. The client's patience, or limited patience, is the key to career risk.
Based on regression is GMO's seven-year forecast. We assume everything will be normal in seven years--profit margins and P/Es--and this unfortunately is what we get: Minus 2.3% for U.S. large cap for seven years in real terms. At the end of seven years, if everything is normal, we get 5.7% real again. But in order to get yields up and returns up, we have to take a hit, because the market is overpriced. Just remember, it tripled in the last six years. It's not unlikely after a triple that it's going to be a little overpriced.
Emerging is the least awful at 3%. Bonds, of course, are unspeakably bad, and timber is hard to access for most people.
We do a diagnostic on every one of these asset classes. This is the S&P 500. On the left column, it says the P/E normal is 16.0. You can make a case that it's 14, but we're using 16. And the current is 21.1, and to take 21.1 down to 16.0 will cost you at the bottom 4% a year for seven years, and then the pain is finished.
Profit margins have averaged 5.7%, they are currently 7.3. To take them down to average will cost you 3.5% a year.
Then we start to get very, very friendly, and the capital growth, which is normally 2.9%, we have awarded a bonus because of high return on equity, and so on. The effective yield 1.7% is what it is, and it sums up to minus 2.3%. So when someone comes on and says it's going to be 3%, ask them to fill out this kind of grid and see how they get there.
Somehow I got this past our legal beagles. This is an actual forecast that we made in December 2000, almost at the peak of the market. You see three columns from the right. We had emerging at 11.4% real for 10 years and at the bottom, we had the S&P at minus 0.8%. The reason I am allowed to use this exhibit is because it isn't cherry-picked, it happens to be the one with the widest gap between two equity asset classes.
So what we were saying back then was that there was a 12.2 [percentage point] compounded difference between emerging markets and the S&P 500. Just think what French and Fama would say to this. They were correlated at 0.6. By the time it finish, they were correlated at 0.7, 0.75. How can you have highly correlated assets moving away this much? You may not know how much 12.2% will do in 10 years. It will take $100 in the S&P and turn it into $320 in emerging. And as you can see, what actually happened was not a 12.2 gap, but a 15.9 gap, so we underestimated it.
But this is the point I have for you. How many small-cap value managers do you have to pick with a couple of points alpha, if you're lucky, in a zero-sum world, to make a dent in two of the five great asset classes on the planet that deviate by 3.2 times. And it was obvious. Emerging in 2000 had taken a beating. It was cheap. It had a low P/E. The S&P was incredibly expensive, 30 times earnings. This was not rocket science. It's a very simple, durable way of betting on regression to the mean. Emerging markets regressed upward, S&P regressed downward.
This, in fact, is the summary of every forecast we have ever made. And I'm not using it to brag; I am going to use it for quite the opposite reason. Although we got the general pattern very nicely correct, you will notice if you had the time, that the average forecast by GMO was 3.3% lower than the actual. So, over on the right, when we estimated 12 to 14, it came in at 15.2. In the middle when we estimated minus 4 to minus 2, it came in at plus 2.2, and the average is we were pessimistic by 3.3%. Why?
Well, 2% of that 3.3% is the fact that we end expensive today, so that is completely consistent. 2.5 points more than I have to explain is because the profit margins in this era have been abnormally high and have not regressed in the normal way, an issue that we will be discussing. And the third factor to balance the books is that, of course, since the market was higher than we thought, the yield was lower by almost 1.3% to balance the equation.
This is the next problem. Why has regression slowed down? Why has everything become sticky? My belief is it has a lot to do with the regime shift from the old days of the Fed to the Greenspan era and his acolytes, Bernanke and Yellen, who follow the identical battle plan. You can see very clearly that the P/E in the new regime for 25 years has been 60% higher than it was for 100 years before. This is not an insignificant shift. This has had enormous consequences. This had made bulls look good and clairvoyant, and bears look slightly idiotic.
And this is what happened to profit margins in the Greenspan era. They used to be 5.9%. For the last 10 years, 8.4%--40% above [the average]. And remember, these multiply out. You have a 60% move in P/E, and then the E is going up by 40. So here it is to make life easy, the P/E markup needs a 37.5% decline to take care of it. The margin markup needs a 30% decline. Multiply those out, you get a 57% decline required to take the data we're looking at in the new Federal Reserve regime and take it back to the prior regimes. Just remember that the prior Fed regimes were Republicans, Democrats, and everything in between. That was a pretty robust long-term average.
How does this compare to GMO's forecast? GMO's forecast minus 2.1 is now minus 2.3 could be taken care of by a 42% decline tomorrow; we would be back to 5.7% real. So, we're looking for a 42% decline; it takes a 57% decline to normalize for the current Fed regime. Therefore, GMO is paying a new-era premium of fully 34%. In other words, compared to the old regimes of Volcker and all his predecessors, we're paying 34% more. So we are compromising, rightly or wrongly, with some of the reasons for higher profit margins and possibly some of the reasons for higher P/Es. So GMO's position is far from being as bad as it could be.
Why are margins sticky? I think it has to do with two factors. One of them, I owe to my friend Andrew Smithers in London, widely quoted in the Financial Times and The Economist and widely ignored in the U.S. He created this idea of the stock option culture. We've entered a world where 80% of the remuneration comes in bonus and stock options, mainly stock options. If you back up 30 years, it was 20%. 80% in stock options combined with the dropping off of the stakeholders and a fixation on short-term profit maximization will do what is happening--and that is: For a senior management person to use the cash flow as a corporation to buy stock back is much less dangerous than building a new plant.
When I was first in the business, I was an analyst, let's say, of paper stocks. And just at the moment when there was a nice shortage developing and profit margins looked like they would zoom upward, every paper company would open two or three new paper plants and crash the market. Back in the '60s and '70s and early '80s, the U.S. corporate system was driven by market share. Everyone wanted to dominate their industry. This was brilliant for capital spending, brilliant for job creation and wage creation. It was terrible for profit margins.
When you build a new plant, there are moments of great risk. You can destabilize a plant as you're shaking it down. A raider can come and get you at a moment of weakness. How much more comfortable it is to buy the stock back, mop up any sellers, push the stock price up, and make your stock options more valuable.
Think for a minute how this interacts with the Fed regime. The Fed has artificially depressed interest rates lower and lower until they disappear. They've made debt incrementally cheap, and therefore other things being even more profitable for corporations. They've made it desperately appealing to borrow cheap debt to buy your own stock back.
We are today buying stock at a world record $700 billion annualized rate year-to-date. Even the low interest rate and the Federal Reserve regime, I think, has tended to make capitalism a little sticky, and this ability to buy your own stock back and have it overwhelm your earnings, I think, has a lot to do with this exhibit that shows how dismal capex is today. Capital spending is four points below average after a six-year economic recovery with world record profit margins.
If you said that to anyone 10, 15, 30 years ago, they would have said, "oh, well, we'll be way above average. We'd be six points higher than we are today of capex as a percentage of GDP," which is a huge amount. This is a very high price to pay to make stockholders in the corporation rich, senior officers rich. It is really a drag on economic growth.
Let's have a quick review of asset bubbles, and see how these things have been affected. This is an infamous exhibit that we dragged up in '98-'99 when we were bleeding like pigs, defending conservatism against the biggest bull market in American history. It had never sold above 21 times earnings; it went to 35.
These are the bubbles that we were able to lay out in front of our clients. We found 28 major bubbles. These are four of each kind. And what we found is that of the 28, all 28 had gone all the way back, they had all broken completely. It was not 23 to 5; it was 28-nil. Every bubble that mattered broke completely.
And a really unexpected feature is that they went back half a year quicker than they went up. Let me debate French and Fama for a second. They say, of course you have bubbles. We define bubbles as 2-sigma events. You have the data, you have the volatility, you can see what 2-sigma is. It's the kind of event, 2-sigma, that occurs every 44 years in a random world. French and Fama say, "well, of course, you have a number of these, because the world changes." And so it does.
But what happens at the peak of these bubbles? In a French and Fama world of market efficiency they go off randomly in all directions. From the peak of every bubble, half of them go up and half of them go down. It's a random world. The peak is a fair price. Some of them go sideways. But what happens in the real inefficient world? Every 2-sigma event--which actually occurs every 30 years in the real world, which is close enough for government work--is followed by an equal and offsetting 2-sigma bear market event, and that should happen routinely every 30 years times every 30 years. Every 900 years you should randomly expect a massive shift in random fortunes of war, followed by an equal and offsetting precise wipeout of the same fortunate event. And how many times has this occurred? 28 out of 28. It is an absolute death knell to the concept of market efficiency and a huge underpinning to my belief and the crazy behavior of Homo sapiens.
A quick update on U.K. housing. This is the housing bubble going up to '96. The Brits do very nice housing bubbles; they obsess about it. They had a very nice one in 1972; look how regular that one is. They don't get better than that. And then they did a slightly better one in '87, and by '96 they were sitting there at three times family income. And then they pulled this ridiculous state of affairs. A 3-sigma classic bubble not followed by a lovely wipeout, but followed by kind of random dithering around, which makes mean reversion look a bit sick. So what is going on here? I will give you my hypothesis.
You get mean reversion if capitalism is allowed to work in the normal way. What happened in the U.K. is they had not built as many houses as they had family formations since 1982. Just think about that. How crazy can you possibly get? That's 33 years they have been allowing immigration at one end of the government and preventing zoning at the local government. The same exact thing has happened in Sydney and most Australian cities. They allow immigration; they don't allow building, and the house prices rise. You can destroy any capitalist mean-reversion process if you interfere with the rules and regulations enough.
This is in comparison to the U.S. housing market--the ultimately painful bubble. And it's why GMO got lucky on the great financial crash, which I believe was really better described as the great housing and resource crash. We saw this happening on the way up. We had complete confidence that it would break, and we got lucky. It broke beautifully. You don't get better behaved bubbles than this one. And because we focused on this, we got it right, and the people who focused on the infinitely complicated financial system tended to get it wrong. No bankers called the crash, and we had a very easy time all the way through late '06 and '07 predicting the death and destruction.
Why? Because the U.S. housing market, if it did what it did, was going to wipe out $10 trillion of perceived wealth. And we knew quite a bit about junkie subprime instruments. We didn't know a third of what was out there, but we knew quite a bit. We knew that if $10 trillion was echoed around by lousy subprime instruments, you were going to have, as I said, the failure of at least one major bank, etc., etc.
The difference between this bubble and the miserable one before in the U.K. is, they didn't build houses, but we built houses like they were going out of style. 1.5 million extra houses over trend each year for three years, and pretty soon you're drowning in houses, and it's only a question of time.
If you've been to Ireland, you know what happened there. House prices went up much more than in the U.S., and they built much more. They were building more houses than the U.K.--a country with seven times as many people. The countryside is littered still with half-finished houses. The whole southern part of Spain is covered in new buildings. They broke the back of their spectacular housing bubble.
So if you don't interfere, you build houses and you break it. This is very important when you get to bubbles, because I believe that what's happening to profit margins and the stock market bubble this time amounts to interference with the normal capitalist process.
We are not allowing profit margins to mean-revert. It used to be the most dependable mean-reverting series in all of finance. It is, in fact, how capitalism functions. If you make an abnormally high return, you are meant to suck in competition. If you make a miserable return, you are meant to frighten it way.
And what has happened this time? For 20 years, we've had abnormally high returns and instead of sucking in competition, all we are doing is buying back our own stock. We are not even responding. This is the equivalent of Australia and the U.K. in the housing market. We are not expanding our economy because of the highest profit margins, like we did for 200 years. We are running away from it and protecting our stock options. So there is no arbitrage mechanism for the time being, and unless we correct this through regulation, it will be an increasing drag on our economy, and we will not have the growth that we expect.
The pressure on U.S. growth is fairly legendary. We have lack of capital spending. Population is down 1%. Income inequality. Everything is adding to an odd situation. And if you listen to Krugman and the boys, what they are saying is, "Productivity is so low. We don't understand. We are confused." Mainstream economics is confused by the fact that we're having chain-link disappointment in economic growth. I'm happy to say I've been writing for six years … that we couldn't expect a growth rate in the U.S. of more than 1.5%. Two years ago, the Fed was saying 3%--completely impossible. And time after time, we are now being disappointed. In September, the Fed said 3.1% for this year, and last week they knocked it down to 1.9%. So the process is going on. They are not very quickly learning what's up.
How is the bubble playing out so far? These are the two most dependable indicators of value: Tobin's Q in yellow, and in black is Shiller P/E, which is 10-year P/E. This is where we were a year ago, and that's where we are today. We are creeping nicely, but slowly, toward 2-sigma land, toward bubble land. I don't believe there is any chance of this bubble breaking until we get over 2-sigma.
We need to trigger to break a bubble. Broad overvaluation has never done it. There have been three broad overvalued markets in history: 1929, but it had a trigger in the U.S. The U.S. equity market was a wonder of crazy speculation, and it was a 2.5-sigma event, and it broke the world equity market.
The next one was 2007, which I described as the first truly global bubble--actually it was the second. Broad overvaluation of all the equity markets, not a single one was a real 2-sigma classic bubble. But what we had, the ace up our sleeve, was a trigger in the U.S. housing market--completely sufficient to the job.
Now we have a broadly overpriced world again. Most equity markets are overpriced. Bond markets are more overpriced than they have been in history. Fine arts are just breaking all the records you read about two weeks ago. These are all necessary conditions, but not sufficient, and a 2-sigma is necessary but not sufficient. What we need is a trigger.
I think what we will have to do is wait until deals become more of a frenzy and individuals become crazy buyers. This, by the way, is the confidence of institutions--pessimism at the bottom. There were no pessimists left a year ago, nor today.
This is the pessimism of individuals. They are pretty optimistic and they are even more optimistic today, but they are not buying stock. That was a year ago; that's today. They are buying the normal amount of shares. No bubble has ever broken until individuals pour into the market.
I believe this market is going to follow the line of least resistance from the Fed and keep going, at least until the election, plodding slowly but steadily higher, waiting for speculation from individuals and from deals.
And the most important single thing probably in looking at the bubble is how middle-aged still this economy is. Great bubbles break when the economy is screaming overrated capacity. 1929, it's just had the most amazing year; industrial capacity went up 14%. 2000, it's over capacity. Japan, it's overcapacity. 2007 it's at capacity. But now it's not. The participation rate of the labor force has come down 4%. There is at least 2% of that which we can scrape back into the workforce. So we have 5.7% unemployed plus 2, plenty of labor.
This old chart again, capital spending. We can have a boom; it's four points down. By the way, private firms that don't have stock options are spending capex just like they used to. It's public firms that are cutting back, which is completely compatible with the Smithers' theory of what is holding us back.
Some people think the trigger will come from a rate increase. Everyone is in a state of mild hysteria at the thought of a single rate increase. Well, let us go back deep into the Middle Ages to the very last such event. The market crashed, hit a low in late 2002, and almost immediately, they raised the rates. They raised the rates from early 2004 until early 2006. They raised the rates eight times, by 4 points, and the market went up all the time without missing a beat, remember? 2004, 2005, 2006. Why, therefore, would a single rate cut have everyone in subject conniption fit this time?
So what is the trigger? You have to wait for the trigger. You have to wait for more oomph in speculation. You have to wait for higher prices. So be braced.
By the way, the Fed came to help in 2000. Immediately it started to come down. And in 2007 immediately they raced to help. The moral hazard from the Fed is profound. If there is a bull market, you're on your own. If there is a bear market, we will immediately leap to your help. And this creates very long seven-year, six-year bull markets, followed by 18-month declines. It's perfect for stock options: You rise, you make a fortune. They crash, and you rewrite the stock options.