원문 보기
On the first working day of the year 2000, an investor named Howard Marx sent his clients a memo called bubble.com. One company had just risen 698% on its first day of trading. Yahoo was worth more than General Motors and Ford together. And Markx wrote that calling it all too high was like standing in front of a freight train.
About 10 weeks later, the NASDAQ peaked and went on to lose almost 4-fifths of its value. The strange part is that Markx had been warning about that market for years and in the same memo he summed up his own caution in a single word which was wrong. He still says that nobody can tell when a market will turn himself included.
So how did he get this one right? The answer is his idea of the pendulum which tells you not when a market will turn but where it stands. In the next 15 minutes, you will learn to take a market's temperature from three readings anyone can look up and then see it measured on one real boom and bust with his own memos from the time beside the numbers.
The idea is older than Markx and it starts with Benjamin Graham who was born in London in 1894 and grew up in New York to become a Wall Street investor who also taught at Colombia. And between 1929 and 1932, by his own account, his partnership lost about 70% of its capital. In the spring of 1949, he published the intelligent investor.
And in it, he asked the reader to imagine owning a small share of a private business. One of your partners, he wrote, named Mister, "Market is very obliging indeed. Every day he tells you what he thinks your interest is worth and sometimes he lets his enthusiasm or his fears run away with him. Graham's student Warren Buffett retold the story to his own shareholders in 1987.
And his summary is the line most people remember. Mister market is there to serve you, not to guide you. So the price a market quotes each day carries a mood as well as a value. And a mood is something you can observe. In August 2024, Markx opened one of his memos with Graham's passage, and he tied it to an image he had first written down more than 30 years earlier, which was a pendulum.
Howard Marx grew up in Queens in New York, the first in his family to earn a college degree. He studied finance at the Wharton School, finished business school at the University of Chicago in 1969, and went to work in the research department of First National City Bank, now Citab Bank, where he rose to director of research.
In 1978, as he later recalled, he got a call from his boss who said, "There's some guy named Milin or something who works for a small brokerage firm in California and asked him to find out what high yield bonds were. Few people had ever heard of them and Markx wrote years later that the brief conversation changed his life.
He moved to a firm called TCW in 1985. And on the 12th of October 1990, he sent his clients the first of his memos titled The Route to Performance. In 1995, he and four colleagues opened Oak Tree. And he kept rioting for 10 years without a single reply until bubble.com. Much later, Warren Buffett said that when memos from Howard Marks arrive in his mail, they are the first thing he opens and reads.
His second memo dated the 11th of April 1991 and titled simply first quarter performance holds the image he would return to for the rest of his career. The mood swings of the securities markets he wrote resemble the movement of a pendulum. Picture it swinging between two ends with euphoria at one end and depression at the other.
Or in his words between overpriced and underpriced. The midpoint of the ark is where the pendulum sits on average. But as Markx pointed out, it actually spends very little of its time there because it is nearly always on its way from one side to the other. The swing also feeds itself because as he put it, it is the movement toward an extreme itself that supplies the energy for the swing back.
So the further the mood runs in one direction, the more there is to come back. When he reprinted that passage 13 years later, he said he would not change a word. Yet, in the very same memo, he wrote that predicting the swings of the pendulum was an unrealistic expectation. So, if a man who writes about cycles all the time says he cannot predict them, what exactly is he looking at?
Part of the answer is that the pendulum never swings alone. In his memos, Markx describes a chain of linked cycles, and he ranks them by how hard they swing. The economy moves in what he calls a gentle wave. Company profits react to it in an exaggerated way, and credit, which is simply the willingness of lenders to lend, swings between markets that are wide open, and markets that are slammed shut.
Credit matters most of all because as he wrote in 2001, it takes only a small fluctuation in the economy to produce a large fluctuation in the availability of credit. On top of the whole chain sits the market itself, which reacts violently because investor psychology magnifies everything underneath it.
In mastering the market cycle, his book of 2018, he gives chapters to the economy, to profits, to investor psychology, to attitudes toward risk, to credit, and to real estate. And he describes a cycle not as a series of ups and downs, but as a series of events, each of which causes the next. Here is one way to draw that chain as an illustration.
Good news lifts the mood. A better mood loosens lending. Easy money lifts prices and rising prices invite more risk until the same links start to pull the other way. The rest of his answer is in a line he wrote in 2006. We may never know where we're going, but we'd better have a good idea where we are.
He cannot tell you when the pendulum will turn, but he believes you can tell roughly where it is. And he calls that taking the temperature of the market. Here is how that works on a madeup market which is only an illustration using three readings you can observe. The first is valuation. How expensive shares are compared with their own history.
The second is credit. How easily lenders hand over money and how little extra interest they ask from riskier borrowers. The third is mood. How confident people say they feel when they are surveyed. In our madeup market, shares cost far more than usual for each dollar of earnings. Lenders are competing to lend on easy terms, and surveys show people feeling unusually confident.
None of those readings says what happens next month. But together, they place this market well out on the greedy side of the ark. And that is all the thermometer claims to do. Markx turned this into a checklist in the same memo of 2006, which he called the poor man's guide to market assessment. It sets out pairs of opposite descriptions such as lenders eager or reticent, spreads narrow or wide, investors optimistic or pessimistic, and asset prices high or low, and it asks you to tick whichever one fits.
If most of your ticks land in the hot column, his advice was to hold on to your wallet. If the pendulum is real, then a past boom and bust should show it in the numbers. So, let's look at one with every date on screen and nothing from today. Here is a closed window of history from January 1997 to December 2009 with three readings drawn month by month from real data.
for valuation. The ratio the economist Robert Schiller made famous known as cape which divides share prices by 10 years of average earnings after inflation. For credit the extra yield that companies at the lower end of investment grade paid above 10-year Treasury bonds and for mood the University of Michigan's monthly survey of consumer sentiment.
Start with valuation because from 1881 to 1996 the Schiller ratio averaged about 15 and in December 1999 it reached 44.2 almost three times that and the highest reading in the whole window. The mood survey peaked 1 month later in January 2000 at 112, its own highest reading of the window in the very month that bubble.com went out.
After that memo, the NASDAQ kept climbing for about 10 more weeks and peaked on the 10th of March 2000. And by October 2002, it had fallen about 78%. 25 years to the day after sending the memo, Markx wrote that it had two things going for it. It was right and it was right fast. But look at the credit strip because it tells a different story.
The extra yield companies paid over treasuries was at its narrowest of the whole window in April 1997, almost 3 years before the stock market's peak. And by the time shares topped out, it was already wider. The three readings did not reach their extremes together. And as the next boom shows, they would not do it the next time either.
After the crash, the pendulum swung back. And this is where the numbers get interesting. Valuation and mood both reached their highest readings of the second boom as early as January 2004 at about 27.7 and 103.8, well short of their extremes of 2000, and neither went higher before the next crash. This time, the heat was in credit.
The extra yield on those bonds fell to its narrowest of the second boom in February 2007, 1.55 percentage points. And that is the same month mark stated a memo called the race to the bottom. In it, he wrote that there was a race to the bottom going on, reflecting a widespread reduction in the level of prudence on the part of investors and capital providers.
In July, he went further in a memo called It's All Good, calling it the strongest, most heated upswing he had witnessed and adding that the tide cannot come in forever. The S and P500 reached its peak that October. He admitted later that he had started complaining too soon, which is exactly what you would expect from someone reading position rather than timing.
The thermometer said hot for a long time before the turn came. And what it got right was the kind of heat lending that had become far too easy. Then the pendulum swung the other way and hard. On the 15th of October 2008, a month after the collapse of Layman Brothers, Mark sent a memo called the limits to negativism in which he wrote that the last few weeks had witnessed the greatest panic he had ever seen, a time when no scenario was too negative to be credible.
All three readings agree with him, though not in the same month. Consumer mood hit its lowest point of the window in November 2000 and ate at 55.3. The credit gap was at its widest in December at 6.01 percentage points, almost four times its width in early 2007, and valuation touched bottom in March 2009 at 13.3, below its long run average of about 15.
His point that month was that skepticism cuts both ways. Skepticism calls for pessimism when optimism is excessive, he wrote. But it also calls for optimism when pessimism is excessive. And that is the pendulum's lesson read from the fearful end of the ark. So what does the window teach? First, the pendulum showed up in every reading which swung from a greedy extreme to a fearful one and back again.
Second, the readings never turned together because in the first boom credit reached its greedy extreme almost 3 years before shares did and in the second it came last. 3 years after valuation and mood. Third, the spacing was uneven. From one greedy extreme to the next took 49 months for valuation, 48 for mood, and 118 for credit.
So no single clock would have fitted all three. And that is exactly why Markx talks about where rather than when. Used as a checklist, the readings let you describe roughly where a market sits on its arc. whether prices, lenders, and mood all look hot, all look cold, or disagree with one another. When they disagree, that tells you which link in the chain is running hottest.
As credit was in 2007, while valuation had cooled from its extreme of 2000. But the pendulum deliberately leaves one thing open, and that is exactly the thing cycle analysis sets out to measure. Markx describes position on the ark. While cycle analysis asks a different question, which is how the swings in price itself have been spaced, how long the rises and falls have tended to last on a given time frame, and whether that spacing has held steady or drifted.
Here is a madeup price series, which is only an illustration with its main waves drawn beneath it. The pendulum reading tells you whether the market is running hot or cold. And the waves tell you how long its recent swings have lasted and how far through the current one price has come. So one gives you the setting and the other gives you the rhythm.
As we saw in the episode on the two Hursts, JM Hurst never claimed that market waves are perfectly regular either because his principle of variation says each wave stretches and shrinks around its typical length. Markx put the same thought his own way when he wrote that the details don't repeat but the rhyming patterns are extremely reliable.
A few cautions keep this honest. It is one window of history measured with monthly averages. So the exact months would shift a little with daily data or a different window. And three readings are only a small sample of what marks watches. A hot reading can also stay hot for years, as it did from 2004 to 2007.
So, the thermometer tells you the conditions, not how long they will last. And as Mark says himself, the cycles he describes aren't predictable as to timing or extent. None of this is a signal to buy or sell. The pendulum describes where a market has stood, which helps you notice when it deserves a closer look.
And what you do with that is always your decision judged against your own strategy. So the next time a market feels unstoppable or hopeless, take its temperature before you trust the feeling. Check how expensive it is against its own history, how easily money is being lent, and how people say they feel, and you will have a fair idea where the pendulum sits.
Even though nobody, Howard Marx included, can tell you when it will swing back. And once you know where a market sits on the pendulum, the next question is where it sits inside its own measured cycle. Want to know how to read that? That is a coming episode on this channel. Subscribe so you don't miss it.
Comments 0