Print Friendly and PDF

The Bear's Lair: Tail risks are the ones to watch

March 10, 2014

In a piece "Dinosaur tail risks" published on Jan. 27, I discussed the various unlikely things that could go wrong with the world economy, mentioning well into the second page the faint possibility of a global war, since the global geopolitical system appeared as fragile as that governing before 1914.

Well, that didn't take long. While the risk of all-out war is still no more than modest, it has definitely emerged as a lot more than a "tail" of less than 1% probability that risk managers usually (but wrongly) ignore.

There is a clear lesson here, for investors, bankers and managers. Risks that are in clear view are analyzed to death, can be managed for, and very often don't eventuate. Conversely tail risks, not provided for in conventional risk modeling and not properly analyzed, are far more likely to cause havoc, although naturally we never know which individual risk will rise up and bite us.

To a great extent, tail risks can be reduced by good management. There was little risk of a major worldwide depression with accompanying collapse of asset prices under the pre-1914 Gold Standard, nor under the Paul Volcker tight-money regime of the 1980s. With real interest rates safely positive, there was little danger of asset prices spiraling into a speculative bubble, causing massive misallocation of investment. Only in the 1920s, when the global monetary system had been dislocated by world war and an accompanying pile-up of debt, did speculation rise to excessive levels in the U.S. economy. Even then, the Great Depression could have been avoided with competent monetary and fiscal management in the U.S., as occurred in Britain.

Similarly there was little risk of default on OECD government debts before 2008, except in Japan. Debt levels were well below 100% of GDP in most countries and even in countries where debts were higher, such as Italy, recent budget policies had been sufficiently conservative to reverse the upward trend in Italian debt seen in the 20 years before 2000. Japan was the only worry; there the government's response to a prolonged semi-recession had been two decades of Keynesian "stimulus" spending and it was already obvious before 2008 that the pause in spending that had occurred under Junichiro Koizumi in 2001-06 was beginning to be reversed.

In the United States, the surpluses of the late 1990s had been replaced by deficits, and there were clear signals—such as the 2002 Farm Bill reversing 1996's reforms—that Congress had become much more irresponsible on spending. However, the deficits overall were still small enough to cause only modest worries and the U.S. debt level was moderate unless you took a truly apocalyptic view of likely future trends in medical spending.

Geopolitically, the risk of nuclear war declined rapidly after the successful conclusion of the 1962 Cuban Missile Crisis, although it reappeared again for a few years in the early 1980s after weak U.S. policies had led to Soviet overreaching. After 1990, the risk of major war appeared to have disappeared, but terrorism was always a hidden threat, as evidenced by the 9/11 attacks in 2001. Since that time, a combination of misguided foreign policy in the West and renewed aggression in several of the West's antagonists has made the danger of a "big war" greater than at any time since 1983. The Ukraine crisis, which may or may not be resolved by peaceful means, is only the latest incarnation of this threat.

The normal attitude of policymakers and risk managers to tail risks of this kind was perfectly encapsulated by the British "Ten Year Rule" of the 1920s, incorporated into official policy in 1928 by Winston Churchill as Chancellor of the Exchequer. This said that defense budgets should be set on the basis that the British Empire was unlikely to be engaged in a major war in the next 10 years. Ramsay MacDonald as prime minister wanted to abandon this rule as early as 1931, but was overruled by his foreign secretary Arthur Henderson and his penny-pinching Chancellor of the Exchequer Philip Snowden. In the event, the rule was abandoned only in March 1932, after the Japanese occupation of Manchuria had made it clear that 10 years was altogether too long a window to be reliable. Even at that, the Cabinet cannot really be faulted for complacency; the Manchurian crisis bore little threat to British interests, while March 1932 was almost a year before Hitler came to power and four months before he first achieved major success in German elections.

Nevertheless, history was to prove that the Ten Year Rule was in force for three years too long, and it was blamed by the Churchill school of opinion for Britain's inadequate rearmament before World War II, even though Churchill himself had institutionalized it. There were likewise a number of statements made by Ben Bernanke and others in 2006-07 that the subprime mortgage crisis, already apparent, was unlikely to lead to major loss. The reality is that both economic and geostrategic history change quite quickly, and that over a 10-year period almost anything can happen.

As investors, there are a number of implications of this. We should never buy bonds or other investments with maturities beyond 5-7 years, because we don't know whether the borrowers will still be there in 10 years' time and, in a fiat money system, we don't know whether the money in which the bonds are payable will be worth anything.

In practice, before 1933 investors thought they could get round the latter problem by buying bonds denominated in gold, but President Roosevelt's 1933 outlawing of gold clauses, disgracefully ratified by the U.S. Supreme Court in 1935, proved that even in say 1927, a year of complete economic calm, prosperity and universal world peace, a 10-year gold-denominated bond of a first-class borrower could not be relied upon. (British investors would have been OK with such an investment, as it turned out, even though 1927 was nowhere near so calm and prosperous a year in Britain as in the U.S.)

Under this principle, while in 2007 a cautious investor might have avoided only Japanese 10-year bonds because of that country's high debt, today such an investor would avoid all 10-year government bonds except maybe those of Switzerland (German bonds would be embroiled in any euro mess.)

Conversely, investors will make no money protecting against a fashionable worry such as global warming, though they may be able to invest profitably in the government-subsidized scams that purport to address it. Similarly, there is probably little to be gained currently from protecting against a crash of the largest Western banks; everybody is aware of the problem and has taken steps to address it.

The conclusion is that investors have underestimated the percentage of their portfolios that should be used to protect against the more unfashionable tail risks. Most conventional portfolios carry protection only against well-analyzed risks like bank failure and global warming that are almost certainly over-provided for. However, as this week's events have demonstrated, portfolios need protection not just against fashionable risks but more especially against unfashionable ones.

Risk managers need to carry out a survey of every possible unlikely risk that may occur, and then test that survey with their more conventionally minded colleagues. Risks that evince a sober nodding of the head and pursing of the lips are probably not a problem; the market is already aware of them and has priced them in. The more dangerous risks are those that when proposed cause conventional opinion to question the risk manager's sanity. Postulating a Russian invasion of Ukraine was sufficient to get Sarah Palin branded as a kook on Saturday Night Live in 2008; by 2012, it was still sufficient for the bien-pensants to raise doubts about the sober Mitt Romney. Today, much less than 10 years after Palin first raised it as a problem, it is a reality. Palin's approach to off-beat risks, and its equivalent in the economic sphere, is the one risk managers need to cultivate.

As for risk management at the macro level, the first imperative is to avoid policies that turn dormant risks, such as German revanchism in 1928, into live ones, such as German revanchism in 1933. Budget deficits and Keynesian spending programs need to be eliminated, because over time they inexorably raise the risk of government default, with all the economy-destroying costs of that event. Monetarily, the brakes need to be kept at least half on at all times, in order to reward capital accumulation and saving and discourage speculation and leverage. And geopolitically—but then you know the appropriate geopolitical risk-avoidance policies, which are beyond the purview of a financially oriented column.

Six weeks. That's all it took for a dormant risk to become a live one. We must mark the occurrence and order our priorities accordingly.


Commentary Archive