Tuesday, November 11, 2014

Kill your investing gurus: Jim Rogers

Jim Rogers is another investing guru who's become famous as much for his personality as his ideas.

Rogers worked for George Soros's hedge fund during its early years, and after leaving Soros in 1980, he took two years-long around-the-world car trips. Each trip spawned a travelogue, Investment Biker for the first and Adventure Capitalist for the second, in which Rogers offers readers a Master of the Universe's view of foreign cultures. According to The New York Times, "One problem is that Mr. Rogers's writing is unexceptional and pedestrian. Another is that he is overimpressed by the depth of his insights."

While the books don't show him developing much curiosity about the places he visited, they do show him developing his public persona: whereas he was once merely a rich investor, today he's Jimmy Rogers, the straight-talking, globetrotting adventurer with a pink bow tie.

An Amazon reviewer for Rogers's more recent book Hot Commodities writes that "Rogers goes for a funny and folksy Warren Buffet style but can't quite pull it off," which is a sharp observation. While Buffett the showman has arguably eclipsed Buffett the investor, the latter remains an accomplished figure. With Rogers, there's no brilliant investor for the showman to eclipse. His public predictions have been very wrong.

It may seem unfair to criticize him for that, since even the best investors often make bad predictions and bad investments. Soros has whiffed his share of predictions over the years, and that hasn't stopped him from being a phenomenon. But the problems with Rogers's investing strategy go beyond a handful of bad calls.

First, Rogers practices what one might call the "stream of anecdotes" style of macro investing. The reasoning behind his investments is mostly ad hoc: in an interview he might say he's bearish on the US because its government is running large deficits, then he'll say he's bullish on Bolivia because he visited the country recently and its cellphone reception was better than he'd expected. (This was his actual rationale for investing in Bolivia in Adventure Capitalist.) While he has strong opinions about everything, those opinions don't amount to a coherent system in which different kinds of information are weighted according to their importance.

Second, two of his most vocal recommendations have been to buy commodities and short Treasuries, and I think his logic is flawed on both counts.


Rogers has been bearish on the US dollar and US government debt for at least the past fifteen years, and probably longer, although the rationale for his bearishness has changed over time. For many years, trade deficits and fiscal irresponsibility were going to generate a financial crisis. More recently, he's claimed that quantitative easing will generate an inflationary crisis.

Here's his opinion from 2003:
The dollar is not just in decline; it’s a mess. If something isn’t done soon, I believe the dollar could lose its status as the world’s reserve currency and medium of exchange, something that would lead to a huge decline in the standard of living for U.S. citizens like nothing we’ve seen in nearly a century.

Here's last year's opinion:
All that money printing has Rogers bearish on U.S. Treasury debt. He said he's shorting government bonds and that if it's indeed the end of the 30-year bond bull market, those shorts will pay off. In particularly he said it's time to short long-dated U.S. government debt.

Notably, Rogers also said it was time to short long-dated U.S. government debt in 2005, when Steven Drobny interviewed him for Inside the House of Money.

Rogers's arguments have an intuitive appeal: everyone knows that large trade deficits and government deficits can't go on forever, and there's widespread skepticism of the Federal Reserve and its actions. These are valid concerns, so I don't mean to dismiss them blithely, but I think the financial crisis that Rogers expects will be neither as immediate nor as inflationary as he suggests.

First, having a reserve currency makes persistent deficits not only possible but likely: in order to accumulate reserves, other countries have to run current-account surpluses with the country that has the reserve currency. Its deficits are a byproduct of reserve currency status, not necessarily a sign that the status is about to end.

And while people tend to think of the United States's deficits as the result of Americans living beyond their means, to a great extent the arrow of causality flies the other way. China and many other countries try to grow by suppressing domestic consumption and subsidizing export-oriented industries, which produces trade surpluses with the US.

Second, quantitative easing isn't synonymous with money printing. (Note: that isn't an endorsement of QE.)

Third, in countries that finance themselves by borrowing in their own currency, there's a tendency for high total debt levels to suppress interest rates. If rates rise, the debt load becomes unmanageable, which pushes the economy into a recession and renews downward pressure on rates. Credit Bubble Stocks has called this idea the self-limiting hypothesis. We've seen this play out in Japan, and I think it will play out in the United States too.


In 2004, Rogers published Hot Commodities, which advocates buying physical commodities as investments. The book endorses the idea of a commodity supercycle, stating that "stocks and commodities have alternated leadership in regular cycles averaging 18 years."

I'm not familiar enough with oil or farm commodities to generalize about their performance, but I've traded metals and mining stocks for a while, and my experiences with them make me very skeptical of the commodity supercycle idea.

There's a subscription site called The Chart Store that offers a wealth of historical charts. For widely-used metals, it has charts going back to late 1800s. Most of these charts show no trend in real prices. The price of aluminum has fallen significantly over the past century, but copper, lead, zinc, and nickel have shown little real change over time. Despite that, they've been quite volatile: metals price cycles tend to be the same length as overall economic cycle, and sometimes they're even shorter.

The mix of volatility and mean reversion means that the typical metal trades in a wide range. For instance, copper has normally traded between $1 and $3.50 in today's prices. On a few occasions, metals have overshot or undershot their range as a group. During the Great Depression, prices plunged far below the low end of their range. The same thing happened in the late 1990s, when many metals approached their Depression-era prices. By contrast, in 2006-07 prices broke above the range's high end.

So commodity prices started the 2000s near record low levels, and less than a decade later they were making record highs. Prima facie this looks like evidence of a supercycle, but that kind of price gain had actually never happened before. It was an unprecedented move rather than a manifestation of a price cycle that commodities repeatedly, inevitably go through. Moreover, there are specific identifiable reasons for both the record lows and record highs, so we don't need to invent the idea of a supercycle to explain the move.

In the 1990s, there was a series of emerging markets debt crises, the most prominent of which were the Tequila Crisis, the Asian Financial Crisis, and Russia's 1998 default and devaluation. The common features of these crises were that they 1) fostered a need for foreign currency, since the crises were generally the result of external debts and 2) crushed the local currency, making commodity production much cheaper. Essentially, the crises provided both a carrot (lower production costs) and a stick (the need for hard currency) for flooding the commodities markets.

During the same period Russia was recovering from the Soviet Union's breakup, and its legal regime was weak and unstable. Most of the people who seized control of newly-privatized companies in 1990s Russia did so through dubious means, and they realized that in the same climate their companies could just as easily be seized from them. This lack of ownership security gave them an incentive to maximize short-term profits. For mining and oil companies, that meant producing as much as they could as quickly as they could. Russia is a major commodity producer, so this dynamic led it to dump large amounts of commodities on the world market throughout the '90s and particularly after 1998.

By 1999, metals were near Great Depression levels. There was a bounce in 2000, but prices fell again in 2001 and 2002 and retested the lows. Yet five years later, many were at record highs.

The obvious cause of this price rise is China's (mal)investment spree and the associated demand for commodities. But financial speculation also played a role, as Frank Veneroso argued in his 2007 World Bank presentation.

Commodities prices surged in the 1970s, and this figures prominently in the commodity supercycle story. But Veneroso points out that the '70s aren't comparable to the more recent commodity price surge because inflation was so much higher in the '70s. To a large extent the commodity price gains during that period merely kept pace with inflation, with the real gains staying within historical norms. Commodities had similar nominal price increases during the 1970s and the 2000s, but the 2000s witnessed much larger real increases.

Veneroso also points out that the rise in commodity prices during the 2000s coincided with explosive growth in both the trading volume and outstanding notional value of commodity derivatives. Some mining executives have also commented on this speculation:

"There's a fair amount of financial money in commodities today," [Phelps Dodge CEO Steven Whisler said in 2006]. "Is that creating some froth? Yes. How much, I don't think anybody knows."

Despite price rises that would normally be associated with shortages, Veneroso claims, companies like the copper fabricator Nexans had no trouble obtaining metals, suggesting that financial demand rather than real demand was fueling the rise.

A few years ago, Howard Schultz from Starbucks made the same claim about speculation in the coffee market. According to a 2010 Reuters article:

"Starbucks Corp Chief Executive Officer Howard Schultz decried the commodity market on Wednesday, saying financial speculators, not product shortages, were to blame for recent price spikes in coffee."

More recently, there are reports of large amounts of copper being hoarded at Chinese port warehouses.

Veneroso devotes much of his presentation to refuting the bull arguments for commodities. He claims that these arguments, while superficially plausible, are really just the kind of new era arguments that are used to justify every bubble.

Whither commodities?

I think the mining industry is heading for a perfect storm in which: 1) Chinese demand will fall as they stop building empty cities, 2) there will be a supply glut as the industry finishes bringing enormous amounts of new capacity online, and 3) financial demand will disappear and metals stocks that have been hoarded will flood the market as prices fall. There will be a huge sell-off that pushes prices below the marginal cost of production and keeps them there for years, and that will put the commodity supercycle theory to rest.

Most commodities have fallen this year as the dollar has rallied, some significantly, and that's already led to a lot of skepticism about the commodity supercycle. So to many readers, the idea that commodities are mean-reverting and have a terrible outlook will be familiar if not obvious. Nonetheless it's worth repeating, because while commodities are down year-to-date, they're at risk of falling much more over the next few years. Ignore short men who tell tall tales about commodities: they are not a sound investment.


  1. He also invested in Fab Universal and joined their board (!!!).

    1. Thanks, I'd forgotten about that. I think that shows the flaw of having an excessively top-down investment style and ignoring company-specific risks. (It also shows he's a doofus.)

  2. http://www.creditbubblestocks.com/2014/11/good-theory-of-commodity-supercycle.html


  3. Excellent summary.

    This is and has been my long-term take on Rogers and China, for what it is worth.

    Which country is more desperate?

    1. The USA exports dollars produced "at essentially no cost" (Bernanke's words) to China.

    2. China exports actual goods produced "at a somewhat higher cost" to the USA.

    I'm going to say that China is more desperate, but maybe that's just me.

  4. For what it is worth, this is what I wrote back in 2010.

    June 3, 2010
    Cumulative Trade Deficit Nightmares

    It would be easy to say that hyperinflation/stagflation is coming when looking at such a model, especially for those of us who live in the United States. I did call myself Stagflationary Mark after all. There's certainly a decent chance.

    I don't think it is that simple though. This is similar to the economic model of the Great Depression. The world was built on the exponential growth of production capacity then too. When the exponential growth in demand is no longer there to meet it we could just as easily slide into an ongoing deflationary spiral.

    You think China is safe? Just look at that chart. Their economy depends on exponential growth that is based on the exporting of goods in exchange for paper dollars. That's not a sign of strength. It's one of desperation. When that began to fail, they opted for a housing bubble of epic proportions to plug the hole. In my opinion, it is making the situation even more dire.

    1. I certainly can't disagree with that, particularly the last paragraph. I think we'll see a an exponential trend in exponential trend failures over the next 10-20 years.

  5. Great post!

    Michael Pettis' "exorbitant burden" article [1] changed my thinking about trade deficits. I assume you've read it, but I didn't see it linked above.

    I've been thinking about whether the "savings imbalance" theory applies as much to generations within a country as much as between countries. For example, one way to interpret China's "excess savings" is that the Chinese people's desire to save exceeded the rest of the world's capacity to find profitable investments, so ever more marginal enterprises get funded (e.g. housing in the US or Spain). If the Chinese knew what assets their retirements were being secured by, they'd probably prefer to spend more now and less in the future.

    It seems like a generalization of the above is that there's a "carrying capacity" to intertemporal savings. US baby boomers or Chinese approaching retirement may wish to continue spending at X% (where X might be ~80%) of their present rate of consumption for a considerable period of time. However, that desire can only be satisfied by creating an ever-expanding universe of promises from the future. Specifically, the only way for retirees to represent Y% of the population (where Y might be ~33%) and continue spending at X% of their working-age rate is for the economy to create a set of promises which annually yield something proportional to GDP * X * Y.

    If you looked at the world, or individual countries, as small scale closed system (e.g. 100 people on an island) it becomes pretty obvious that future consumption expectations among the soon-to-be-retired are completely unrealistic. You can't have a politically stable island society where 45 working people support 20 children and 35 retirees, if the retirees demand to spend at a high fraction of their working-age rate. That's tantamount to bondage. Yet, that's exactly what many developed economies are trying to work out how to do.

    There's just no way 35% of the population can afford to drive around in Winnebagos.

    I think a lot of debt, such as student loans and mortgages, can be thought of as a post hoc rationalization for the transfers people have imagined for themselves in the future. The debts have nothing to do with what borrowers can reasonably afford to part with, and defy all standards of fairness (they are not voluntary), and yet every time the system attempts to correct itself the status quo is reinstated through ever more extreme measures (e.g. higher taxes, more debt, higher asset prices).

    Most people, including people like Buffett, seem to think they can offer future generations a a one-sided deal and everyone will accept it an behave the same as before (e.g. maintain historical rates of household formation), but that's not how people work. If you institute a bunch of effectively Malthusian policies on working-age people they don't work harder to produce the same number of children, they have fewer children.

    [1] http://blog.mpettis.com/2014/10/are-we-starting-to-see-why-its-really-the-exorbitant-burden/

    1. Thank you for the excellent comment. You're right re: Pettis. He's a brilliant theorist and his first book, The Volatility Machine, has influenced me more than any other book I've read. I've been meaning to do a review of it, but it will probably take a while to write.

      To be honest, I haven't thought much about the intergenerational dynamic, but that sounds right, in the sense that any conflict between savers and debtors will have a built-in conflict between old and young. I also think you're right about the unrealistic assumptions, it seems to be something that few people appreciate or are willing to face, but they'll have to face it relatively soon.

  6. Berkshire to Buy P&G’s Duracell for $4.7 Billion of Stock

    It's hard not to see many of Buffett's recent acquisitions as nostalgia-driven. Car dealerships, retail banks, real estate brokerages are all great investments if the next half century resembles the last one.

    It concerns me that Buffett is betting quite heavily on inefficient intermediaries which don't really add much value. All of the examples above rely on political power to make themselves essential intermediaries - there's no fundamental reason for them to represent such a large part of the economy anymore. The same could be said of buying a railroad which transports a lot of oil - makes perfect sense if no one is allowed to build new pipelines.

    1. In this case he may be getting a good deal. It looks like he's paying less than Gillette originally paid:



Note: Only a member of this blog may post a comment.