玩火（可能冲向旧高点的竞赛）

GMO QUARTERLY LETTER April 2010 Playing with Fire (A Possible Race to the Old Highs) Jeremy Grantham It’s spring, and this spring a young man’s fancy lightly a lot in over 10 years and on our data is likely to have a turns to thoughts of speculation. The Fed’s promises look second consecutive very poor decade, but we have had good and, as long as you’re not a small business, you can two wonderful recoveries in which the more speculative borrow to invest or speculate at no cost. The market has you were, the more money you made. So why not break had a near record rally, sprinting far past our estimated fair the historical rules and try a third time? Perhaps this time value of 875 for the S&P 500. Bernanke is, in fact, begging it will be lucky. us to speculate, and is being mean only to conservative investors like pensioners who cannot make a penny on Still, it does seem inefficient for the Fed to help us up their cash. Collectively, we forego hundreds of billions and then lead us off the cliff again. And to do it twice of potential interest, but at least we can feel noble because seems like sadism. And for us to play the game once we are helping to restore the financial health of the banks more seems like lining up behind hot stoves and begging, and bankers, who under these conditions could not fail to “Please, can I burn my hand a third time?” Investors used make a fortune even if brain dead. We are also lucky to to be more pain averse. It used to be “once bitten, twice have a tiny fraction of our foregone interest returned by shy.” This time, surely it should be “twice bitten, once the banks as loan repayments with “profit.” Some profit! bloody shy!” The key shift seems to be the confidence Oh, for the good old days when we could just settle for we now have in Bernanke’s soldiering on with low rates a normal market-clearing rate of interest. But that, I and moral hazard to the bitter end, if necessary, cliff or suppose, would be wicked capitalism, and we had better no cliff. The concept of moral hazard has changed. It get used to bank- and speculator-benefiting socialism. used to be a vague expression of intent: “If anything goes wrong, I will help you if I can.” It seems to have been The massive bailout program stopped the meltdown of transmuted into a cast-iron commitment. The Fed seems the financial system and engineered at least a temporary to be pledging that it will bail us out after every flood. All economic recovery. We know the obvious cost of this that is lacking is a rainbow! bailout: unprecedented deterioration of the Federal balance sheet. But what of the less obvious costs incurred by Speculators are not stupid. They see that after each crash, taking away the rewards of caution by saving the reckless a long, artificial period of low rates and easy financial and incompetent? These weak enterprises, financial and borrowing has been delivered. They see that Bernanke is an other, were not gobbled up by the stronger, more prudent, unreconstructed Greenspanite in that he refuses to address and more competent natural survivors, and there is a long- bubbles, but will leap to help ease the pain should a bubble term cost in that. break. With asymmetry like that, why not speculate? And so another bubble appears and then another. This time, the So now, Bernanke begs us to speculate, and we are recovery for the total market was 80% in one year, second obedient. Despite being hammered down twice in 10 only to 1932, and the really speculative stocks are almost years and getting punished for speculating, we again double the market, as they also were in 1932. But frankly pick ourselves up off of the canvas and get back into the 1932 was far worse than our crisis where, according to our good fight. Such persistence is unprecedented – 20 years research, only 7% of the market value of speculative stocks for each really painful experience has been the normal remained, compared with 35% this time. Back then, they recovery time – but Uncles Ben and Alan have treated us deserved that kind of rally. And even though I guessed last so well in these two disasters that, with hindsight, they April that we would have a quick rally to 1100, this looks don’t feel so bad after all. Yes, the market is still down quite likely to be far more.

I’m convinced that this excessive market response has In that world, the market would have to decline, but occurred because stocks are far more sensitive to both low not disastrously, and would probably exercise no really rates and the Fed’s promises than is the economy. The damaging effect on the economy. economy is limping back into action, but faces some tough long-term headwinds that I collectively call “seven lean If, however, the economy only limps along, which seems years.” Mortgage defaults in housing, steady repayments of more likely to me, then we run a very real danger of a third consumer debt, and refinancings in commercial real estate dangerous bubble in stocks and in risk-taking in general. and private equity, are all problems that linger, as do many For in that event, Bernanke will definitely keep rates low others, on what is becoming a long, boring list.

We may quarter after quarter and speculation will surely respond. get very lucky and have a strong broad-based economic Again? Yes, I’m afraid so. In that environment, Bernanke recovery. The economy’s durability and flexibility is will do nothing to let the air out gently. His lack of anti-usually undersold by the bears, and I have generally been bubble action is pretty much guaranteed. The end of leery of underestimating its potential. But we can probably such events is always hard to predict, but usually bubbles agree that the economy is plagued by unusual problems this break for almost any reason when they are big enough. Of time. It is therefore perhaps more likely that the economy course, the larger the asset bubble, the bigger the shock to will recover in fits and starts, and that over several years it the economic and financial system.

Now, Greenspan was will underperform its historical record. lucky enough to inherit Volcker’s good work, and that gave him a base from which he could launch or blow a huge If the economic recovery is slow and if unemployment equity bubble; he also had the advantage that the country’s drops slowly, then Bernanke will certainly keep rates very balance sheet was in excellent shape. Even Bernanke low, as he has promised in as clear a way as language inherited a reasonably solid position from which to fund permits. In that case, stocks and general speculation will a second bailout. But a third time? It is hard to work out very probably rise from levels that are already overpriced. where the resources would come from to resuscitate the And if they do, Bernanke will definitely not be concerned economy if a real shock were to be delivered by another and has told us as much.

There were some teasing collapse of a major asset class. The key problems here are comments from Bernanke at the lows last spring to the the Fed’s refusal to see the risks embedded in asset class effect that the Fed might take the embedded risk of asset bubbles and the willingness of both the Administration class bubbles more seriously, as many foreign central and Congress to tolerate this dangerous policy. Heck, bankers have begun to, and very sensibly so. But that they recently reappointed him! Yes, the Congressional hope has now been utterly squashed, and Bernanke has natives were restless, but in waiting for a third crisis to returned to the original Greenspan line: let the bubbles kick him out, they may be too late to avoid the major-look after themselves. Even if we were to re-enter bubble league suffering caused by his blind spot. territory in a way that would be obvious to anyone who can tell the difference between 15 P/E and, say, 28 P/E Should unemployment linger at high levels, which I think (35 of us at last count), he still will do nothing.

For he is likely, and I get these things right better than half the is now once again genuinely unconcerned with bubbles time (I believe about 52%), then we had better hope that and even doubts their existence, as proven conclusively something lucky turns up to break the speculative spirit. by his comments during this last one, the 100-year U.S. This is perverse, but so is Bernanke. What could go wrong, housing bubble, the breaking of which landed us in the preferably in the next few months? Some combination of rich and deep manure of 2009: “The U.S. housing market the following: an unexpected second leg down in house has never declined,” etc., etc. No believer in the existence prices and a continued rise in the level of defaults, leading of bubbles could ever say such things. to a crisis at Fannie, etc.; a wash-out in commercial real estate and private equity caused by refunding problems If we get lucky and have a strong, broad, and sustained (along the lines of Goldman’s and Morgan Stanley’s economic recovery, interest rates will probably rise before recent real estate fund wipe-outs) that result in a chain we reach real bubble territory.

As rates rise, the market of major defaults in properties like Stuyvesant Town; a will almost certainly settle down, and we will only have crisis in the euro where Portugal or Spain or Greece, or all to deal with a substantially overpriced U.S. market and three, default and strange things start to happen; a rapid moderately overpriced global equities and risk premiums. rise in commodity prices, despite the anemic growth of GMO 2 Quarterly Letter – Playing with Fire – April 2010 the developed world, which, with the same caveats, I also enough the first time – he stimulated Year 3 as well. The think is quite likely; competitive devaluations leading to result was that we entered Year 3 in October 1998 and a serious trade war; or my colleague Edward Chancellor’s Year 3 in October 2006 with horribly overpriced markets, favorite, two or three wheels falling off of the Chinese and still the market went up, and by a lot.

The overpricing economy, which today acts as the main prop to global in October 1998, by the way, was so bad that our 10-year growth. Okay, enough. We all know that there is plenty forecast was down to -1.1%; in October 2006, by a nerve-that could go wrong. Some combinations would be enough wracking coincidence, our 7-year forecast was -1.0%. If to break the market but still leave the economy limping the market is 1320 by this coming October (up 10% from along. This would be far better than having the market today), our 7-year forecast will again be -1.0%. (Please rise through the fall of next year by, say, another 30% to hum the Jaws theme here.) Do not think for a second 40%, along with risk trades similarly flourishing and then that a very stimulated market will go down in Year 3 all breaking. The possibilities of this happening seem just because it’s overpriced … even badly overpriced. nerve-wrackingly high.

The developed world’s financial So far it has had 19 tries to go down since 1932 and has and economic structure, already none too impressive, never pulled it off. We can, of course, hope that this time would simply buckle at the knees. will be exceptional. Even in the best of times, though, overpricing is only a mild downward pull. Its virtue is And, briefly, let me give you my reasons why this rally that it never quits. Eventually it wears the market back running through next fall is not at all out of the question. down to fair value.

In October we enter the third year of the Presidential Cycle, the year every Fed except, of course, Volcker’s, So what do I think will happen? That’s easy: I don’t helped the incumbent administrations get re-elected. know. We have been spoiled in the last 10 years with Since 1932, there has never been a serious decline in many near certainties – mainly that real bubbles would Year 3. Never! Even the unexpected Korean War caused break – but this is definitely not one of them. Not yet only a 2% decline. Even when Greenspan ran amok and anyway. (However, I am still willing to play guessing over-stimulated the first two years instead of cooling the games despite the fact that “I don’t know.” So here, as system down – which he did twice, having not suffered Exhibit 1, is my probability tree.) The general conclusion Exhibit 1 Probability Tree: The Line of Least Resistance 0.3 0.30 Ec Economy has a strong and sustained recovery, rates rise, o no market falls, but basically all is well m y bu m ps al 0. o ng 7 ,r at es st ay lo 0.7 w 0.49 No real market shocks, speculation and market prices rise P to October 2011 to dangerous levels, then soon break with m oor severe consequences o av n ec oi ths on di ng bre om lo ak ic d 0. ng s a 3 er ani ta o -te m r rm al cri m spir sis aj it in or s, bu ma nex bb rk t fe le et w s fa lls 0. , 21 Source: GMO Quarterly Letter – Playing with Fire – April 2010 3 GMO is that the line of least resistance is a market move in the managers like us to ever overweight an overpriced asset, next 18 months or so back to the old highs, say, 1500 to so we struggle on the margin to find kosher ways to own a 1600 on the S&P, accompanied by an equivalent gain little more emerging in order to give them the benefit of the in most risk measures, followed once again by a very doubt. I recommend that readers do the same. The urge to dangerous break. If that happens, rates will still be low weasel and own a little more emerging is a direct result of and thus difficult to use as a jump starter, the financial the lack of clearly cheap investment alternatives. system will still be fragile, and the piggybank will be more or less empty. It is remarkably silly for the Fed to allow, Odds and Ends even encourage, this flight path. It is also remarkably silly for investors to be so carefree, given their recent 1) SEC and Goldman: to those who said that hedge experiences. Fortunately, there are several less likely funds and proprietary trading had nothing to do with outcomes that collectively, I hope, are equally probable. the crisis, this recent SEC charge speaks for itself. We are definitely playing with fire and need some luck. Watching hedge fund players both outside and inside The best kind of luck would be that Bernanke gets bitten their banking firms making billions of dollars was an by a Volcker bug. obvious seduction to everyone. It led individuals and even firms to become more aggressive in risk-taking and in interpreting the codes of ethical behavior, and Recommendations Goldman is probably no worse than average. The real Our policy is simple: however complicated the world issue here is more about ethical conflicts with clients may be, we will play by the numbers. The global equity than about legal restraints. These were, in any case, markets taken together are moderately overpriced, and the mostly disassembled by the last four administrations. U.S. part is now very overpriced but not nearly so bad as If we want to be serious about regaining reasonable it could be. Surprisingly, within the U.S. the large high standards of client protection, then hedge fund-like quality companies are still a little cheap, having been left proprietary trading should of course not be allowed totally behind in the rally. They are unlikely to do very within banks. well in a bubbly environment, however long it lasts, but 2) The U.K. and Australian housing bubbles may be should be great in declines and in the end should win. unimportant to U.S. investors, but to bubble historians A potential plus for quality franchise stocks in the next they look extraordinary. The U.K. event in particular few years is that they are far more exposed to emerging has broken out of any previous mold. Despite the countries and, as investors fall in love with all things usual cry of “special case,” they will decline around emerging, this should be seen as an increasing advantage.

40%, back to trend, as was the case for the previous A mix of global stocks, tilted to U.S. high quality, has 32 bubbles. If not, it will be the first time in history a 7-year asset class forecast of about 5% excluding that a bubble has not behaved in this way. Reversion inflation compared with a long-term normal of about 6%. to trend will involve considerable pain, which I will Not so bad. On balance, therefore, we are only slightly discuss further next quarter if things are quiet. underweight equities.

3) Attached is the first half of a short and accurate letter Within my personal portfolio, I have a stronger preference on global warming by the heads of both the National for the already overpriced emerging market equities than Academy of Sciences (U.S.) and the Royal Society do my colleagues at GMO, and actually more than I should (U.K.). Couldn’t have done better myself! have as a dedicated value manager. This is because I believe they will end up with a P/E premium of 25% to 50% in 4) I also include here a link to a video of my April 19 a few years, as outlined two years ago in “The Emerging Financial Times interview about bubbles, which saves Emerging Bubble” (Letters to the Investment Committee me a whole section of writing. It is also a testimonial to XIV, April 2008). The appeal of emerging’s higher GDP talking so fast that they can’t ask you too many difficult growth compared with the slow growth of U.S. developed questions! http://link.brightcove.com/services/player/ countries is proving as compelling as I suspected, and I bcpid71778049001?bclid=69928231001&bctid=7912 would hate to miss some modest participation in my one 8759001 and only bubble prediction. It is hard, though, for value GMO 4 Quarterly Letter – Playing with Fire – April 2010 FT corn E2-14MT I, MANUAL TIMES FT Home > Comment > Letters What's happening to the climate is unprecedented Published: April 9 2010 03:00 I Last updated: April 9 2010 03:00 From Prof Martin Rees and Dr Ralph J. Cicerone. * Sir, We were stimulated by your editorial "Cooler on warming" (April 5). There has undoubtedly been a shift in public and media perceptions of climate change — a consequence of, at least in part, leaked e-mails from some climate scientists and the publication of errors in the fourth Intergovernmental Panel on Climate Change report.

However, as your editorial acknowledges, neither recent controversies, nor the recent cold weather, negate the consensus among scientists: something unprecedented is now happening.

The concentration of carbon dioxide in the atmosphere is rising and climate change is occurring, both due to human actions. If we continue to depend heavily on fossil fuels, by mid-century CO 2 concentrations will reach double pre-industrial levels. Straightforward physics tells us that this rise is warming the planet. Calculations demonstrate that this effect is very likely responsible for the gradual warming observed over the past 30 years and that global temperatures will continue to rise — superimposing a warming on all the other effects that make climate fluctuate.

Uncertainties in the future rate of this rise, stemming largely from the "feedback" effects on water vapour and clouds, are topics of current research. ...

• Martin Rees is President of the Royal Society and Ralph J. Cicerone is President of the US National Academy of Sciences.

Disclaimer: The views expressed are the views of Jeremy Grantham through the period ending April 23, 2010, and are subject to change at any time based on market and other conditions. This is not an offer or solicitation for the purchase or sale of any security and should not be construed as such. References to specific securities and issuers are for illustrative purposes only and are not intended to be, and should not be interpreted as, recommendations to purchase or sell such securities.

The securities discussed in the Financial Times interview are owned by GMO portfolios. This should not be construed as investment advice and is not an offer or solicitation for the purchase or sale of any security. The specific securities identified are not representative of all of the securities purchased, sold or recommended for advisory clients, and it should not be assumed that the investment in the securities identified was or will be profitable. GMO reserves the right to purchase additional shares or sell shares at any time based on market conditions, new information, future events, or any other factor.

Copyright © 2010 by GMO LLC. All rights reserved.

Quarterly Letter – Playing with Fire – April 2010 5 GMO GMO SPECIAL TOPIC April 2010 Letters to the Investment Committee XVI* Speech at the Annual Benjamin Graham and David Dodd Breakfast (Columbia University, October 7, 2009), edited for reading. (Part 2 may follow next quarter.)

Part 1: “Friends and Romans, I come to tease Graham and Dodd, not to praise them.” (On the potential disadvantages of Graham and Dodd-type investing.)

Jeremy Grantham The main struggle I’ve had my entire investment life is with the chapters that mattered to me. What I found surprised the preposterous belief that all information is embedded me; this in particular: “[The] field of analytical work may so quickly and efficiently into stock prices that asset class be said to rest upon a twofold assumption: first, that the bubbles cannot possibly occur. But to be honest, I’ve also market price is frequently out of line with the true value; been pretty irritated by Graham-and-Doddites because and, second, that there is an inherent tendency for these they have managed to deduce from a great book of 75 disparities to correct themselves. As to the truth of the years ago, Security Analysis,1 that somehow bubbles and former statement, there can be very little doubt – even busts can be ignored. You don’t have to deal with that though Wall Street often speaks glibly of the ‘infallible kind of thing, they argue, you just keep your nose to the judgment of the market’ … The second assumption is grindstone of stock picking. They feel there is something equally true in theory, but its working out in practice is faintly speculative and undesirable about recognizing often most unsatisfactory. Undervaluations caused by bubbles. It is this idea, in particular, that I want to attack neglect or prejudice may persist for an inconveniently today, because I am at the other end of the spectrum: I long time … and the same applies to inflated prices believe the only things that really matter in investing are caused by over enthusiasm or artificial stimulants.” If the bubbles and the busts. And here or there, in some ever we were living in a world of artificial stimulus, it country or in some asset class, there is usually something is now. (Also, the great quote attributed to Keynes that interesting going on in the bubble business. The rest of “The market can stay irrational longer than the investor the time, if you keep your nose clean, you will probably can stay solvent,” comes to mind here. Keynes and keep your job. But when there is a great event, that’s the Graham and Dodd agree a whole lot more than I would time to cash in some of your career risk units and be a have thought.) Security Analysis continues, “the market hero. And it turns out that Graham and Dodd themselves is not a weighing machine … Rather should we say were not nearly as anti-the-big-picture as Graham-and-that the market is a voting machine … product partly of Doddites would have you believe. reason and partly of emotion.” More shades of Keynes.

This weekend it dawned on me that I had never read Now, I have heard that weighing and voting machine line Security Analysis. I had very strong opinions about it, but misquoted a billion times by you guys in this room. It is had never actually read it. So I did my best to cover all of not a weighing machine!

1 Graham, B. and Dodd, D.L., Security Analysis, McGraw-Hill, 1934.

• The Letters to the Investment Committee series is designed for a very focused market: members of institutional committees who are well informed but non-investment professionals.

So I have come, friends and Romans, to tease Graham be adequate when reasonable calculation is supplemented and Dodd, not to praise them, even though this is the 75th and supported by animal spirits, so that the thought of anniversary of Security Analysis. And my second point of ultimate loss which often overtakes pioneers” – and nearly attack is that Graham and Doddery is all a little obvious. always overtakes Graham-and-Doddites – “is put aside as I was brought up by a Quaker and a Yorkshireman – that a healthy man puts aside the expectation of death.” You is known as “double jeopardy” in the frugality business. only undertake dramatic initiatives of the type that create Quakers believe waste to be wicked and Yorkshiremen, the Microsofts or Apples of the world with a heavy dose who consider Scotsmen to be spendthrifts, consider it of animal spirits. If you Graham-and-Dodded it, you criminal. The idea that a bigger safety margin is better than would never do anything spectacularly successful. And a smaller one, that cheaper is better than more expensive, this willingness to roll the dice is an important relative that more cash is better than less cash, deserves, in modern advantage for the U.S., and too much risk avoidance will parlance, a “Duh!” It is just rather obvious, and going on simply kill this instinct. about it for 850 pages can get extremely boring.

Let me move on to make a point about how illogical I The next negative point comes from my much admired think it is to leave out the great bubbles and the great busts Chapter 12 of Keynes’ General Theory [of Employment, and focus on the grindstone. That’s my main complaint Interest and Money] – as for most of the rest of Keynes, with you guys: very, very narrow focus. There you are, as far as I am concerned, you can take it or leave it. It is working away, picking stocks, even when the world is vague, contradictory, and sometimes dangerous, although having its occasional cataclysms.

I admire his reintroduction of the importance of “animal spirits” as a potential wrecker of the best laid economic When you buy a stock, because it has surplus assets or a plans (there is a nice new book on the subject by Hunter good yield or a great safety margin, you are really making Lewis2). But Chapter 12 is a pearl, a polished pearl. It a bet on regression to the mean. We are really counting explains how the market works. And along the way, on the fact that current unpopularity will fade, that the Keynes makes the point – he makes a lot of points that current problems in the industry will dissipate, and that cut across Graham and Dodd – that you all here represent the fortunes of war will move back to normal. Well, as a a threat to the economy: Keynes believes that if we had provable, statistical fact, industries are more dependably a margin of safety and showed the typical prudence that mean-reverting than stocks, for individual stocks can Graham and Dodd recommend, no one would undertake on rare occasion, permanently change their stripes à la to initiate a single new enterprise.

Over 80% of all new Apple. (Or is that à l’Apple?) Sectors, like small caps, enterprises have failed fairly quickly in the past. The ones are more provably mean-reverting than industries. The that make it have to struggle with a very uncertain future. aggregate stock market of a country is more provably Graham and Dodd were not at all comfortable with the mean-reverting when mispriced than sectors. And great future. They thought that dealing with it was speculative. asset classes are provably more mean-reverting than a They much preferred the present. What are your assets in single country. Asset classes are the most predictable of the piggy bank now? What is the yield you receive today? all: when a bubble occurs in a major asset class, it is a It’s all quite irrational because they are prisoners of the near certainty that it will go away.

(A bubble for us is future just like anybody else. However many assets you defined as a 2-sigma event, statistical talk for an event have in the corporation, including cash, can all be eroded that would occur randomly every 40 years under normal long before you can get your hands on them. conditions, a definition that is arbitrary but at least to us feels reasonable. And we define a “near certainty” as over Keynes continues, “… if the animal spirits are dimmed 90% probable.) and the spontaneous optimism falters, leaving us to depend on nothing but a mathematical expectation, enterprise For the record, I wrote an article for Fortune published in will fade and die … It is safe to say that enterprise which September of 2007 that referred to three “near certainties”: depends on hopes stretching into the future benefits the profit margins would come down, the housing market community as a whole.

But individual initiative will only would break, and the risk-premium all over the world would widen, each with severe consequences. You can 2 Lewis, H., Where Keynes Went Wrong: And Why World Governments Keep perhaps only have that degree of confidence if you have Creating Inflation, Bubbles, and Busts, Axios Press, 2009. been to the history books as much as we have and looked GMO 2 Letters to the Investment Committee XVI, April 2010 at every bubble and every bust. We have found that there Let me tell you a story to illustrate this last point. In are no exceptions. We are up to 34 completed bubbles. 2000, Gary Brinson ran broad-based portfolios of global Every single one of them has broken all the way back to the assets, as did we. He did it for UBS, then, the largest trend that existed prior to the bubble forming, which is a pool of money in the world.

He rotated his mix around very tough standard. So it’s simply illogical to give up the to avoid troubles and to take advantage of cheaper asset really high probabilities involved at the asset class level. classes. (This seems a perfectly sensible approach but All the data errors that frighten us all at the individual is a very tiny part of our industry.) I considered Gary stock level are washed away at these great aggregations. in the late 1990s completely brilliant. That is to say his It’s simply more reliable, higher-quality data. portfolio looked identical to ours. He was underweighted in stocks and largely out of growth stocks. Conversely, Keynes thought that the Graham and Dodd approach, he was heavily overweighted in value stocks. And two if done in an institutional world, was also incredibly weeks from the market peak, because they had lost about dangerous to your job.

“Investment based on genuine 25% of their asset allocation business as growth stocks long-term expectation,” Keynes wrote in Chapter 12 in surged, he was fired from UBS/Brinson. As was Tony 1936, “is so difficult today as to be scarcely practicable. Dye, a die-hard Graham-and-Doddite who ran a very He who attempts it must surely lead much more laborious value-based contrarian portfolio for Phillips and Drew, a days and run greater risks than he who tries to guess better UBS subsidiary. Gary, by the way, is unlike most of us than the crowd how the crowd will behave; and, given contrarians: he is a capable administrator and generally equal intelligence, he may make more disastrous mistakes made of steel. If any of us could withstand the corporate … It needs more intelligence to defeat the forces of time pressures to go with the flow in a major bull market, he and our ignorance of the future than to beat the gun.” could.

It was a fair test, and had the tech bubble lasted Keynes understood that what really drives our industry, just a month or two less, his bets would have been then and now, is momentum, career risk, and beating the wonderfully successful and we would have had to share gun. “Moreover, life is not long enough – human nature that anti-growth market niche with a real 800-pound desires quick results, there is a peculiar zest in making gorilla. So his firing was very convenient for us. Today, money quickly … The game of professional investment I don’t believe any public company could withstand the is intolerably boring and over-exacting to anyone who rapid loss of business involved in opposing an extreme is entirely exempt from the gambling instinct.” All of bubble on the grounds of overpricing. Management would you here have of course been injected with the Graham simply not stand for the hit to quarterly earnings involved and Dodd anti-speculation serum, so my sympathies for in the inevitable loss of business that comes from fighting the boredom that you have to suffer.

“Finally it is the a bull market. After Gary’s firing, a normally reasonable long-term investor … who will in practice come in for the “trade rag” suggested his stance had been eccentric and most criticism, wherever investment funds are managed moving to a more traditional balance of growth stocks – by committees … For it is in the essence of his behavior despite their being at 65 times earnings – was, all things that he should be eccentric, unconventional and rash in considered, less risky. Less risky, that is, for the manager’s the eyes of average opinion.” Average opinion, by the next quarter's business, not less risky, of course, for the way, is prudence. Prudence is defined as doing what a ultimate beneficiaries, the pensioners. similarly well-educated person would do. Therefore, if you are not going with the pack, you are imprudent.

Sorry Meanwhile, back in Boston, we, unlike UBS, had no hand guys, all of us contrarians are, by this standard, imprudent. holders and no marketing people then. And in our asset To continue with Keynes: “If [our value manager] is allocation division we lost 60% of our book of business. successful, that will only confirm the general belief in his We lost more than any other competitor that we are aware rashness ….” (I like to say that when he’s successful he of, then or now. And we lost it by making the right bets will be patted on the back but, when he leaves the room, for the right reasons - bets we ultimately won. It was a he will be described as a dangerous eccentric.) “[And] if wonderful hothouse experiment – a perfect demonstration … he is unsuccessful … he will not receive much mercy.” to prove Keynes’ hypothesis.

And we lost the business The pure administration of Graham-and-Doddery really quickly – in two and a half years. In the fall of 1997 we had needs a long-term lock-up, like Warren Buffett has, or it a good several-year record in asset allocation, and two and will have occasional quite dreadful client problems. a half years later we had lost 60% of the book of business!

Letters to the Investment Committee XVI, April 2010 3 GMO To be more serious in my criticisms, a potential weakness gets higher and higher as its price goes down. These of the Graham and Dodd approach, as it is usually companies almost always end up going down less than the practiced, is in its reliance on low price-to–book (P/B) average stock. When there is a really severe recession, ratios as one of its cornerstones. Low P/B ratios are, after however, the dividend starts to get cut and it becomes a all, the market’s way of saying “these are the assets in little more questionable. And when there is a depression which I have the least trust.” It should not be surprising, or a crash, then the companies start to get cut – to go out therefore, that when you have a depression, or nearly have of business – and “value” companies get to take serious one, that more of these “cheap” companies go bust than is pain.

We sent someone into the stacks to get data from the case for the “expensive” Coca-Colas. These serious 1929 to 1932 (he nearly died of dust inhalation). This economic setbacks can give us serious value traps. We data (Exhibit 1) is completely proprietary and it must be had one starting in late 2006, where cheap companies said that some contradictory data has also been dug out of became cheaper and cheaper and quite a few ceased to the archives. If this data is correct, as we believe, then it exist. And several more that were blatantly bankrupt were certainly shows the hidden risk of low P/B. I think P/B bailed out by the government for reasons that still seem and yield and price-to-earnings (P/E) are risk factors. quite arbitrary and desperate rather than capitalistic. With They have less fundamental quality and are therefore a less corporate-friendly government, the loss involved in more prone to failure in rare crashes.

I think this is the one this value trap would have been far worse. In my opinion, thing Fama and French got right – for the wrong reasons. despite the pain taken by many heroes of the Graham and On everything else, of course, I disagree with them. Dodd world, you were still collectively desperately lucky, saved by the Great Bailout. Exhibit 1 shows the number of times your holdings had to increase from 1932 to get back to the 1929 level. If you The other value trap that was impossible – or improbable were expensive, on the left, you had to go up 6.4 times. – to avoid was the Great Crash. Normally, a cheap But the cheap stocks with the best P/B ratios had to go up company with lots of assets and a high yield outperforms 14.3 times to get their money back. Too many of them in a bear market because it’s propped up by the yield that had gone the way of all flesh.

Let’s assume we get two Exhibit 1 The Hidden Risk of Low Price/Book Stocks – Price/Book in the Great Depression 14.3 Multiple Needed to Break Even 11.7 8 6.4 6.3 5.4 Expensive 2 3 4 Cheap Price/Book Quintile Post Post 1933 1933 expected expected risk risk premium premium Time Time required required to to catch up up of low low Price/Book Price/Book stocks stocks with with high high Price/Book Price/Book stocks stocks 2.0% per year 41 years Source: GMO GMO 4 Letters to the Investment Committee XVI, April 2010 points a year for the extra fundamental risk of carrying sort of approach – buying a handful of names that he cheap P/B stocks. That 1932 drop chewed up what really understood. He became very suspicious of the amounts to 41 years’ worth of a reasonable risk premium! idea that diversification could be an advantage. It just That was the value trauma of the century.

The rest of the meant he argued, that you owned a lot of stocks you didn’t time until 2007, admittedly with temporary interruptions understand well. It really sounds like Buffett, doesn’t or ebbs and flows, you made extra money buying low it? And he became a contrarian. Quote: “The central P/B and low P/E. But in 1929 you basically took such a principle of investment is to go contrary to the general hit that you had a hard time getting back out of the hole. opinion, on the grounds that if everyone agreed about its Let me take this opportunity to point out, courtesy of Jim merits, the investment is inevitably too dear and therefore Grant, that Ben Graham lost 70% in the Crash. That’s unattractive.” So, ironically, Graham and Dodd are less 70% of his clients’ money. He went into the Crash highly Graham and Doddy than you like to think, and Keynes, leveraged, net long, apparently completely unaware of the the Father of Momentum – the beauty contest, musical possibility of a speculative bubble about to burst.

The chairs, and the quick draw – is much more akin to the great value manager, master of the safety margin, was traditional view of Graham and Dodd and Buffett than is more than 100% long equities! No wonder by 1934 he commonly thought. was very, very conservative. That will do it! (And by the way, just to rub it in, Roy Neuberger went into the Crash The “cheapest” P/B ratios have another potential net short; that’s a big head start.) weakness. Sometimes they are not usefully cheap at all. The range of P/B ebbs and flows to a magnificent degree Incidentally, Keynes too got wiped out in the early 1920s, as shown in Exhibit 2. In 2000, the range between the currency speculating, and was bailed out by a rich friend. P/B of the market favorites and the market pariahs was That’s fine if you’ve got rich friends.

He didn’t do that very, very wide. As wide as it had ever been. When the well later on in the Crash either, but he began, in the range is wide, the top end – the high P/E favorites – are early 1930s, to get the point. He had been hammered very vulnerable, and the cheap, contrarian stocks at the enough that he began to adopt a rather Warren-Buffetty other extreme can make you a fortune. The top exhibit Exhibit 2 Even for Price/Book and Small Cap, Relative Value Is Very Unforgiving 3.0 Price/Book* +18% Overpriced 2.5 “Death of Value” Relative Strength 2.0 1.5 19½ years to break even 1.0 0.5 0.0 Dec-70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 00 02 04 1.9 Small Stocks +22% Overpriced 22 years & counting 1.7 Relative Strength 1.5 1.3 1.1 0.9 0.7 Dec-70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 00 02 04 * Best 25% price/book by name ** Stocks 600 on by market cap Source: GMO As of 9/30/06 Letters to the Investment Committee XVI, April 2010 5 GMO here shows a peak in 1983, when I am very pleased to counts.

What this means is that any outperformance on say I gave a talk in Boston called “The Death of Value.” our intrinsic value is pure alpha, where for P/B, etc., and It was looking like a crowded trade. Everybody wanted for small cap it is a risk premium, and a risk that definitely to be a value manager by 1983 because it had done so comes to bite you every so often. Yet the client world has dazzlingly well since 1974. It had beaten the market by seldom been interested in this apparently vital difference, over 100 percentage points! The growth managers were which is an interesting commentary on where our industry hiding under the table. Yet from 1984, because value has been on this issue. Outperformance of a benchmark investing became so trendy, you made no extra money is usually everything, and risk-adjusted returns nothing. in the cheapest P/B (value stocks) for 19.5 years!

Now For us, this approach has been a disadvantage. For the that takes patience! You were paid absolutely nothing industry, it has pushed managers into ignoring risk in extra for carrying the lower fundamental quality that P/B value management. represents. Exhibit 3 shows, relative to the market, this To cut to the chase, P/B does not represent intrinsic extra risk that P/B derives from being very low quality. value. Nor do P/E ratios or yields. To make this point I Quality here is measured in the standard GMO way, using regularly pose a question to investment audiences: “I give principally the level and stability of profitability and you Coca-Cola at 1.2 times book or General Motors at secondarily the level of debt. This exhibit also shows the 1.0 times book. Hands up, who wants General Motors?” similarly low fundamental quality of small cap, so it also No one ever puts up their hand, and I say, “Therefore, is a risk factor.

The final bit of data on Exhibit 3 is GMO’s Q.E.D., P/B is not value.” You know that the extra intrinsic value series, which recognizes that quality qualities represented by Coca-Cola are worth a premium. and growth deserve a premium. On this basis, half the The question is only, “How much?” time Coca-Cola is “cheap,” and half the time expensive, while Microsoft spent several years in the best decile! The simple “value” measures outperformed nicely in Traditional value that wants its assets and yield now would the good old days, probably for three reasons. First, never score the great companies as cheap. Yet they must they represented the higher fundamental risk shown have been for they outperformed, which is the only check above – a higher risk of commercial and financial failure. on the accuracy of historical value measures that really Second, they represented higher career and business Exhibit 3 GMO Value Has S&P Quality – and It Is High Today More Quality GMO Value 7 High Quality Today S&P 500 Deciles of Quality 4 Small Price/Book More Junk Dec- 70 73 76 79 82 85 88 91 94 97 00 03 06 Source: GMO As of 9/30/09 GMO 6 Letters to the Investment Committee XVI, April 2010 risk.

It is hard to justify having bought a contrarian, by divine right, regardless of how they were priced. unpopular stock when things go wrong, which happens These factors in the past had delivered the goods because quite often. Even reasonable committees felt it was an the “spreads” – the range between large and small cap obvious risk and only imprudent managers would have and between high and low P/B ratios – had been wide. bought it. In contrast, when a Coca-Cola has a bad time As they became mainstream “risk factors,” and with the in the market, the same committee tends to see it as a popularity from their huge success in the 70s, the ranges sign of the market’s superficiality in not recognizing the narrowed. When the range between Coca-Cola and U.S. stock’s great characteristics.

This extra career or business Steel on P/B becomes narrow, it can still easily be picked exposure should not be borne by value managers without up and modeled but, it will fail to deliver an excess return. the expectation of a higher return. Before the mid-1980s Low P/B stocks, or small cap stocks, only outperform this was, generally speaking, the case, for at that time (and when they are priced to do so, as I hope every Grahamite this is the third and most important reason) the investment knows. community was more risk averse than now so that, with 1929 as the sole exception, stocks with low P/B ratios, low And this was precisely the problem by mid-2006. After P/E ratios, etc., and small caps typically over discounted some strong years of performance, the range of old-the specific problems and the general low quality and fashioned value measures such as P/B and P/E had consequently outperformed. become severely compacted – the range between the low book ratios and high book ratios had been bid down.

Yet This state of affairs in which simple value measures some very illustrious Grahamites, including a couple of outperformed was changed by two events, perhaps well-known hedge funds, were saying that “value” was forever. First, there was the massive outperformance quite well-positioned. of “value” from 1973 to 1983 when the cheapest decile of P/B outperformed the market by over 100 percentage Exhibit 4 shows exactly how the attractiveness of P/B points. Second, a few years later a newly arriving wave ebbed and flowed on our data. The period starting in of statistically well-educated “quants” adopted P/B and 2006 when P/B reached its maximum overvaluation was small cap as winning factors that should be modeled. a pretty shocking time – a 50-year flood for P/B, P/E, and Egged on by French and Fama, et al., they tended to value managers in general.

This was the modern value assume that these “risk” factors delivered an extra return trap from hell, a reminder of 1932.

Exhibit 4 Cheap Price/Book Are Not Always Very Cheap Valuation of Value Stocks* Relative to the Market 0.8 Expensive 0.7 Relative Valuation of Cheapest 25% 0.6 0.5 0.4 On Price/Book vs. Market 0.3 0.2 0.1 Dec-70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 00 02 04 06 08 *Cheapest 25% (of largest 1,000 stocks) on Price/Book Source: Compustat, GMO As of 9/30/09 Letters to the Investment Committee XVI, April 2010 7 GMO Small Caps there any when we started at Batterymarch in 1970 with a Like low P/B stocks, small caps peaked in 1983 (see small cap portfolio.

Exhibit 2), but unlike them, small caps have never regained their old relative high of that year. Yes, small Quality caps have won over the very long run and had a truly A missing ingredient in this critique of Grahamism, or wonderful rally after 1972, but who do you think goes rather Grahamism as usually practiced in the real world, bust in the Great Depression? The big blue chips with all is probably Warren Buffett, whose introduction would those workers to protect, or the little companies? If the conveniently bring up the topic of Quality, which typically governor of some state has one telephone call to make is something of a missing ingredient in value investing. It to the President, he makes it for a Lockheed, he doesn’t is what he really introduces as an extra emphasis into the make it for some unknown little company.

The small caps world of safety margins and attractive traditional value had to go up 14 times to get their money back, the blue measures. chips 6.8 and 5.4 times. Note that 5.4 isn’t a very low multiple, but these were tough times. It’s just a whole lot If the rare value traps are the bane of Grahamism, then better than 14 times. The time taken to catch up if you equally they offer an opportunity for quality stocks to had, say, a reasonable 1.5% risk premium for small caps, show their merits. In Exhibit 5 we show the relative would have been 48 years. Basically, small cap investing performance in the Great Crash of Quality’s close cousin, was brilliant for 60 years, but if you had been managing high return on equity. The high return companies that money in small caps in 1929, you would almost certainly entered the Crash overpriced still outperformed brilliantly. have been knocked out of the game, having dug too big They had a princely 25% of their money left at the low – a hole too quickly.

Would any clients have allowed you whoopee! – whereas the low return firms were left with the time to recover when they were left with 7% of their 5% of theirs so that they had to quintuple just to catch money? I suppose the good news is that there were no up with their high return brethren! If you had picked up small cap managers in 1929; nor for that matter, were a risk premium of 1% a year for holding low quality – Exhibit 5 Quality as Armor Plating Is Free and When You Really Need It – Quality in the Great Depression 20.5 Multiple Needed to Break Even 10 8.5 9.0 7.8 4.1 Low Quality 2 3 4 High Quality Outperformance Outperformance needed needed Number Number of of years years required required ifif for for Low Quality Quality to catch Low Low Quality Quality outperformed outperformed up up to to High High Quality Quality by by 1% 1% per per year 404% 163 Source: GMO GMO 8 Letters to the Investment Committee XVI, April 2010 which on average you had not – it would have taken you academics several decades after investment practitioners nearly 165 years to catch up. at Batterymarch and elsewhere had been using these factors to make money.

On noticing this outperformance, In fact, Quality stocks have outperformed the market embarrassingly late in my opinion, Fama and French since 1965 (when our quality data begins) as shown in adopted a circular argument rather typical of finance Exhibit 6. We define “quality” using primarily a high academics in the 1970 to 2000 era: the market is efficient; and stable return. I think you would agree that this is a P/B and small cap outperform, ergo they must be risk workable definition of a franchise since to be both high factors. That the result in this case happens to get to the and stable means you have the ability to set your own right result is luck. The real behavioral market is perfectly prices. Secondarily, we look at debt. This yields a very happy not rewarding “risk” when it feels like it, as is shown uncontroversial list of stocks of the Coca-Cola, Johnson by the 70-year underperformance of high beta stocks.

But & Johnson, and Microsoft ilk with not even one financial! this time it worked. Price-to-book, despite its low beta, is Even though the “quality” factor is now cheap, it has still a risk factor because of its low fundamental quality and its outperformed by a decent (maybe you’d say “modest”) vulnerability to failure in a depression. This is true with 40% over almost 50 years. But this 40% is an amazing free small cap as well. But what about “Quality?” This factor lunch. Warren Buffett doesn’t really talk much about the has outperformed forever. (The S&P had a High Grade fact that he is playing in a superior universe. Why should Index that started in 1925 and handsomely outperformed he? It’s like having the Triple A bond outperforming the S&P 500 to the end of 1965 when our data starts.) the B+ bond in the long term by 1% a year when, in a Since the market is efficient, to Fama and French quality reasonable world, it “should” yield, say, 1% less.

And must be a risk factor! So, by protecting you in the 1929 how nicely this messes up the Fama and French argument Crash and in 2008, and by having a low beta for that on risk and return. matter, Quality as represented by Coca-Cola and Johnson & Johnson must be a hidden risk factor. Oh, I know: “The That P/B and small cap outperformed was noticed by real world is merely an inconvenient special case!”

Exhibit 6 Quality: Finally, a Free Lunch – High Quality Stocks Win Over the Long Term 70% 60% Cumulative Return of 50% 40% 30% 20% Quality Stocks Relative to S&P 500 10% 0% -10% -20% Dec- 65 67 69 71 73 75 77 79 81 83 85 87 89 91 93 95 97 99 01 03 05 07 Note: GMO defines quality companies as those with high profitability, low profit volatility, and minimal use of leverage. The historical valuation is determined by our proprietary intrinsic valuation measure. Source: GMO As of 9/30/09 Disclaimer: The views expressed are the views of Jeremy Grantham through the period ending April 23, 2010, and are subject to change at any time based on market and other conditions. This is not an offer or solicitation for the purchase or sale of any security and should not be construed as such. References to specific securities and issuers are for illustrative purposes only and are not intended to be, and should not be interpreted as, recommendations to purchase or sell such securities.

Copyright © 2010 by GMO LLC. All rights reserved.

Letters to the Investment Committee XVI, April 2010 9 GMO