Not just a coincidence
Some analysts and investors, reluctant to draw generalisations from a few freak years, place the credibility of this theory on par with believing in the tooth fairy. But the statistics since 1990 show that, on average, spreads do widen and excess returns do drop going into October.
Chart 1 shows the average month-end spread over Treasuries of US investment-grade corporates. Chart 2 shows the average monthly excess returns for the same securities. Both show a clear tendency towards underperformance (there is not yet enough data on the euro-denominated market to make a strong case either way).
This would mean little if the four extraordinarily poor Octobers were merely distorting the averages – but this is not the case.
Certainly, a credit slump does not occur every October: excess return performance improved in the October months of 1993 and 1999, and the Federal Reserve’s injection of liquidity into the market immediately after the September 11 terrorist attacks prompted a huge bond rally in 2001.
But aside from these three years, one of which involved exceptional and, one hopes, never-to-be-repeated circumstances, every year since 1990 has shown some degree of deterioration going into October. In three of the 13 instances, excess returns continued to fall into November, but more often, October has seen the nadir of the second-half slump.
No one knows exactly what is behind this track record of October downturns. For this reason, investors find it difficult to predict whether the October curse will strike in any given year, and if so, to what extent.
One explanation points to the stock markets. Equity has its own October effect, in that it tends to become more volatile in the autumn. Chart 3 shows the daily averages on the Market Volatility Index (VIX) since 1986. Daily averages since 1997, when the Asian financial crisis ushered in a period of volatile second halves, are also shown. Both show a marked increase in volatility going into October, which will have some implications for the credit market.
But most analysts believe that if there is an October effect, it is driven not by fundamental cycles in credit or equity, but by investor spending cycles. “Historically, the first half of the year always pulls in the best returns. By autumn, many investors have hit their targets for the year, and opt to lock in their profits,” says Gary Jenkins, head of credit research at Barclays Capital.
Credit professionals taking their summer holidays means that liquidity tends to dry up in August, reinforcing investors’ caution. This seasonal decline in risk appetite is also exacerbated by the fact that the US fiscal year ends on September 30. At this point, most commercial and investment banks make an effort to get risk off their books.
Perversely, it may well be this caution that makes investors more vulnerable to any autumnal blow-ups. “Investors start thinking about locking in profits as the third-quarter end draws closer,” says Simon Surtees, credit strategist at Gartmore Investment Management. “They hope that it will be a quiet end to the year. But this caution leads them to become jittery, and when some event does rock the boat, investors overreact.”
In other words, if a market catastrophe has already occurred elsewhere, the credit market will become most vulnerable to any knock-on effects later in the year.
A repeat performance
There are several potential developments that could trigger an October slump this year. The possibility of an equity correction in the US market looms over the second half, with early earnings announcements for the third quarter beginning in October. Any unpleasant surprises at that time could make the market volatile. If the economy does not grow at anticipated levels, further interest rate action could also spell trouble. And the recent turmoil in the treasuries market might have unforeseen consequences for credit.
There are sceptics, however. Stephen Dulake, credit strategist at Morgan Stanley, says: “There is little evidence that credit investors are less interested in credit now that many of them have met their targets for the year. All over Europe, portfolios remain at least neutral on credit. The few institutions that have tried to go short on credit have been frustrated.”
Others simply deny that the theory can be taken seriously. “I just don’t think it is possible to predict shocks in this way.” Says Suki Mann, credit analyst at Société Générale.
But Jenkins urges caution: “While the October effect may seem like a coincidence, it would still take a brave man to say it is not going to happen.” Even in the absence of a painful market event, he argues, fears of an October downturn can become a self-fulfilling prophecy. “And the worst part is, it’s not like the investor can say he wasn’t warned. You always know there’s a chance that it will happen.”
The week on Risk.net, July 7-13, 2018Receive this by email