Knowledge Management Initiatives in Singapore

The dreaded September to October, the March to September 2009 rally, and future scenarios

In this third and final chapter on this topic based on a 7Cities/Wilmott talk given on Thursday, 26 February 2009, I discuss the dreaded September- October, the March to September 2009 rally and future scenarios, inflation versus deflation and the US debt bomb, housing trends, what's happening in China and Japan, income inequality and government regulation and the next bubbles.

Historically, September and October have been the worst performing stock market months in the US and across the world. Indeed, the graphs I showed in Sell-in-May and Go Away in the May issue of Wilmott indicated th at the rule sell on May 1 and buy on the 6th trading day lbefore the end of October, for the !S&P500 and Russell2000 futures indices for the 16 years 1993-2009, did indeed handily beat a buy and hold strategy. For the S&P500 a buy and hold strategy turned $1 on February 4, 1993, into $1.56 on August 17, 2009; whereas, sell in May and move into cash, counting interest (Fed funds effective monthly rate for sell in May) and dividends for the buy and hold, had a final wealth of $3.18 , some 103.8% higher. For the Russell2000 , the final wealths were $1.49 and $3.69, respectively, some 147.7% higher for the sell in May strategy. Chapter 17 updates this to December 2011 and the longer results are even better. Among other things, the huge drop and volatility in August and September 2011 was avoided.

As September approached, a distinguished Princeton University professor and very good student of the markets was pestering me regarding whether September- October would be bad months, especially following a huge move up in a weak economic climate with the S&P PE ratio over 100!

He was my guest at Saratoga for the August 28, 2009 Travers race - we did win on another anomaly. In the previous ten runnings of the grade I images

mile Travers, the top 3-year-old race between the classics (Kentucky Derby, Preakness and Belmont and the Breeders' Cup) was won by a winner of one of these classic races. This is similar to grade I class, most grade I races are won by previous grade I winners even if they look terrible in recent other races. There was only one such candidate and he looked good on the handicapping models we follow - that was Summer Bird and indeed he won easily.

These months certainly were dismal in 2008 with September falling 9.79% and October falling a huge 20.11% with November's -9.22% and February's -11.13% completed the decline on March 9. See the VIX volatility index in Figure 25.1. In the fall of 2008 the VIX was approaching an astronomical 90,now it has fallen to 25% in the March to September rally. This September 2009 started out with a small decline as many were worried and sold positions in advance of a possible fall on the first trading day but the market stabilized and rallied in the following days until the Wednesday September 23 Fed meeting. In the Fed minutes, they stated that the loose money policy will continue at very low interest rates for quite some time well into 2010 and there was no immediate danger of inflation. See Figure 25.2 for current US and other interest rates. All this is positive for the markets but the markets interpreted it as a slowing of the slow but steady recovery that has led the market up since the March lows. Figure 25.3 shows the S&P500 and the increase has taken the market from 666 to 1070 with a three-day decline to 1044.38 as of the close on February 25 when this column was finished.

The decline was not much and weathered the storm of a weaker than expected durable goods report, lower US housing sales. When stock markets rally, they carry many other securities and assets along for the ride. This includes thoroughbred stud farms, art, closed end country funds, etc., See Figure 25.4 for the all art index and Figure 25.5 for the closed end funds index. Observe the declining discounts along with the rise in prices. In Kentucky for the Derby and Oaks (the filly Derby) in early May 2009, I visited several top stud farms such as Claiborne and Three Chimneys. Here the correlation is strong; stud fees follow the S&P500 but levered. For example, Fusaichi Pegasus, the only Kentucky Derby winner who was sired by the top speed sire Mr Prospector, had his stud fee of $200K moved to $25K even when he is the fourth leading sire in the US.² Other stallions had declines as well; $75K to $15K for example for the near Triple Crown winner Point Given. Of course, some have stayed constant and a few rose for outstanding performance. These prices are now rallying. Equities have a more significant wealth effect overall since equity market moves tend to influence consumption decisions of the highest income groups, who in turn account for the greatest proportion of US consumption (though out of their income they consume a lower proportion than the poor).

I discuss this more below, but remember that the US debt is in US dollars and many other countries like the UK and Japan have much more as a percent of GDP. See Table 25.1. Their debts are also in their currencies. Before we discuss this rally, let's look at one more data snooping anomaly and then the facts on September- October over the recent past.

This Rosh Hashanah anomaly relates to the Jewish new year. In 2009, that was close of September 18 to open on September 28 for the trading days The origin of this practice seems to be the belief of Jewish investors that they should liquidate their portfolios during the holiday so that their attentions could be fully focused on their worship; or more likely in today's world, not trade.

Going back to 1915, the performance of the DJIA is -0.62% from the last close before Rosh Hashanah until eight days later with the last close before Yom Kippur. From then to December 31 averaged a respectable 1.99%, see TheStreet.com and the Kirk Report.

Table 25.1: Debt as a percent of GP around the world. Source: Laing (2009)

images

Source: Japanese Journal of Population; United Nations; International Money Fund

What happened in 2009. The close on September 18 was 1068.30 and the close on September 25 was 1044.38, so -2.24%. So the anomaly worked once again.

Rosh Hashanah for Jewish Year 5771 occurred on sunset Wednesday, September 8, 2010 - nightfall September 10, 2010 and Yom Kippur began on Friday, September 17, 2010. The close on September 8 was 1098.87 and the close on September 16 was 1124.66, for a gain of 2.35%. So the anomaly did not work in 2010. The close on December 31, 2010 was 1257.64 or a gain of 11.82% from the close on September 16, 2010.

Rosh Hashanah 2011 began on September 28 and the S&P500 cash close then was 1151.06. Yom Kippur 2011 began on October 7 with the S&P500 close at 1164.97. The change was +1.2%. So the anomaly failed again.

Returning to September-October. Figures 25.6ab and 25.7ab show the monthly effect in the S&P500 large cap and the Russell2000 small cap future indices, respectively, from 1993–2011 and 2004–2011. Observe that over the longer horizon October is actually positive, averaging =1.29% per year, and September is slightly negative and there is no reliable monthly effect. As the historically strong months of January and February are negative, December is reliably positive though.

More recently, from 2004–2011, October is massively negative and November, which historically has the strongest turn-of-the-month is also negative. The year by year September-October returns from 1993 to 2011 are in Table 25.2. The pattern is clear: these months are like other months except that they frequently have big declines-8.38% (2001), 11.31% (2002), -9.79% (2008) and -7.85% (2011) for September and the -20.11% (2008) for October for the S&P500. October had other great falls in 1929, 1987 and other years. The Russell2000 is similar. Additional data is in Dzhabarov and Ziemba (2009).

Bialkowski, Etebari, and Wisniewski (2009) study stock returns during the Muslim holy period of Ramadan in 14 predominantly Muslim country during the period from 1989"–2007. They find that stock returns during Ramadan are almost nine times higher (38.1% versus 4.3%) than during the rest of the year and that this conclu-sion persists after controlling for other known calendar anomalies like the January Effect, Halloween Effect, and Turn-of-the-Week Effect. Their explanations for this phenomenon rely on recent behavioral theories that connect investor emotions with their decisions. Specifically, they suggest that these excess returns are a result of increased investor optimism experienced during Ramadan as a time of relative hap-piness, solidarity, and social identity for Muslims; they go as far as to suggest that Ramadan may cause ‘mild states of euphoria,’ as suggested by Knerr and Pearl (2008). This upbeat or positive sentiment then causes relative overconfidence and an increased willingness to take risk, such that investors perceive investments as of relatively higher value. Bialkowski et al. also consider that during this relatively healthy period there may be a higher demand for equities, as documented by Rosen and Wu (2004). They, however, do not find any evidence of a higher trading volume during Ramadan.

In their study, Bialkowski et al. discuss Ramadan to shed light on its potential emotional effects, review the clinical effects of fasting, and review empirical evidence on the effects of Ramadan on equity prices in 14 Islamic countries. Their empirical results compare the average returns during the holy month and the rest of the year and find that 11 of 14 countries studied have higher returns during Ramadan; the countries that did not exhibit this anomaly are Bahrain, Saudi Arabia, and Indonesia. They mention the effects of relatively few observations in the case of Bahrain and Saudi Arabia and note the outlier of the Asian Crisis in Indonesia effecting its equity prices during Ramadan.

They further test their results by two event studies, benchmarking returns against a constant-mean-return model as well as a predictive market model using a proxy of 23 industrialized countries that do not have Muslim majorities. In these tests, they calculated cumulative abnormal returns (CAR) as returns in excess of what an investor should expect in the absence of Ramadan, finding this CAR to fall between 2.5%-3.1% depending on the event study approach. Among other robustness tests, Bialkowski et al. analyze if this CAR may be compensation for increased risk during Ramadan by examining return volatility, but do not find any supporting evidence and in fact find that, except Turkey, all countries studied actually showed decreased volatility during Ramadan.

Bialkowski et al. also test their results for illiquidity effects, exchange rate considerations, and other accepted calendar effects, and find that the Ramadan Effect remains anomalous and can most likely explained by temporal changes in investor psychology. They conclude that investors can profit in Muslim stock markets by buying shares at the beginning of Ramadan and selling them at its end, though they do not explicitly model this strategy nor do they estimate transaction costs.

The big 2007 to March 2009 decline from over 1500 to 666.79 on the S&P500 and similar declines across the world plus all the associated banking, real estate, unemployment and other crises had led to a deflationary environment. To counter act this, the US Fed and Treasury bought billions and billions of securities that were to say of questionable and hard to measure value. They created stimulus programs such as the cash for clunkers, and a $700 billion rescue program for the banks. Meanwhile, the Fed reduced short interest rates to essentially zero. These various actions plus a declining flow of bad news such as monthly unemployment rising but at a declining rate improved the economic outlook. This second derivative effect plus the low interest rates, with huge monies on the sidelines and resumption of investor confidence led a rather large over 56% increase in the S&P500 from the intraday low of 666.79 on March 6th and a low close of 676.53 on March 7th on the S&P500 to the 1440.67 when we went to press on September 30. Meanwhile, the US consumer remains on the sidelines afraid to spend much so for the first time in a long while US savings is positive. This, of course, cuts down on imports from China and elsewhere and thus there is a downward pressure on prices. Hence, there is some deflation and bonds, especially high yield bonds (they are called that or junque instead of junk because of their 50% return in 2009) have been stellar performers, all related to credit risk confidence returning to the market. A good way to measure this confidence is through carry trades where one borrows in a low yield country to invest in a high yield country (like Australia or Canada). The yen exchange rate is one signal of this risking with increasing confidence and falling with decreasing confidence. See Figure 25.8 for the yen valued in US dollars and euros updated to . . . .

Fig. 25.8 The yen versus the US dollar and euro, 5-years to October 12, 2012. Source: Yahoo

The debt built-up from the Fed and Treasury actions amounts to an estimated $11.6 trillion. Some of this money is being paid back by the recipients of the rescue like Goldman Sachs whose stock has risen from a low of 47.41 to its current 183.58. Recall it was Goldman and GE that Warren Buffett's Berkshire Hathaway made loans to at 10% with some free call warrants at 120 for Goldman as part of the deal. Look for Berkshire to move eventually out of its doldrums.

So paying back this debt could well lead to higher interest rates. This week, the noted Tiger Hedge Fund manager Julian Robertson, an astute student of the markets, is expecting this but big time. He sees the possibility of 15-20% interest rates and similar inflation. He cannot predict when but he sees it coming. This is, of course, an extreme view but from such a noted analystit deserves consideration. Consequently, it is one possible scenario. My own feeling is that it is way too extreme and will not happen - but 1981 had such interest rates and 1968-88 when Ed Thorp was running his Princeton-Newport Fund, the S&P returned 10% but interest rates averaged 8% for these 20 years. So Robertson is short long term bonds which seems likely eventually to be a wise trade. It is clear that eventually inflation and interest rates will rise but Robertson's values are way too high since the consequences of such an event are too dangerous to consider. As of March 2012, there are minor signals of higher interest rates in a couple years but nothing major. But the bull market in bonds is close to over.

In inflation, one thinks of gold and silver so lets see how they have done relative to other assets. Figure 25.10 shows that in the past five years, gold has outperformed silver which has greatly outperformed the S&P500, oil and the dollar index.

Fig. 25.10 Comparing gold, silver, oil, spy and dollar index (dx). Source: Yahoo

In Vancouver, where I live, housing is on fire! But it has been overpriced relative to Vancouver incomes all 40+ years I have been here. It is usually people who made money somewhere else who are pushing the prices and sales. Recently that's China. It used to be Hong Kong since Expo 1986 then other Asian places like Taiwan and Singapore were active. Lately it is more mainland China. The winter Olympics are coming in February 2010 but that is not a major factor yet. The prices and sales picked up as the S&P500 and Toronto TSE300 rallied and interest rates are rock bottom: 2.5% variable, for example. So it is easy to buy if you have a substantial down payment. There are no subprime mortgages in Canada nor non-recourse mortgages. In the US, housing is still weak but stabilizing.

Homeowners' equity has declined from 58.7% back in '05 to around 43% today. What's more, nearly a third of households have no mortgages, which, of course, means that the equity percentage of the 50-plus million that do have mortgage loans is a good cut lower than 43%.

Out of 56 million units, 6.94 million are in deliquency and eventually headed for liquidation and 300,000 join the queue each month. So housing is still weak. As well, commercial real estate is much weaker and has not yet hit bottom.

As they said in Indiana Jones, you've got the wrong Jones, I mean Ziemba, on this one. See Rachel's columns in this magazine and at Roubini Global Economics in New York. It has been a couple years since I was last in China so I defer to and rely on Rachel who follows it closely. In a recent article in the USA Today, she discussed some key points, see Lynch (2009). Chinese consumption is slowly on the rise, about $1.15 trillion in 2009 versus $1 trillion in 2008, a 15% increase. But it is less than a quarter of the up to $700 billion US consumers are not spending even if they spend their long run average. Since 2005, the Chinese yuan has risen 16% but still is significantly undervalued. A further rise would boost domestic demand so would making credit more widely available, Rachel argues. The G-10 wants less credit-fueled shopping. Europe needs to invest more and China consume more. The US current account deficit was 6.6% of GDP in 2005, now it is 2.8%. But there is still a huge US deficit and China's current account surplus. Figures 25.11abc show new brokerage accounts, the Shanghai stock exchange and the US dollar/yuan over time. Observe how the growth in brokerage accounts mirrors the Shanghai stock exchange.

The credit extension and improved confidence helped reflate Chinese equities. Concerns that the Chinese government is tightening the credit that helped lead to economic recovery and asset market reflation contributed to a 23% fall of Chinese equities in the month of August, pushing the CSI 310(0, a measure including stocks trading in Shanghai and Shenzhen markets, into a technical bear market. The index rebounded 16.7% in September and J still up more than 80% from .January to September 15. Price/Earning ratios have more than doubled from their lows in November 2008, but remain below their peak in January 2008. Some analysts suggest as much as 30% of the short-term bills financing helped fund equity trades, meaning the decision not to renew bills financing reduced inflows. Other factors contributed to reduce liquidity to the equity market in July and August including the restart of IPOs, issuance of money-market bonds and expiry of share lockups. Meanwhile the number of new brokerage accounts had its 2009 peak at the end of July, right before the Shanghai composite peaked. Thus, Chinese equities seem vulnerable in late 2009, but should be supported by economic growth in 2010.

Figure 25.12 shows the dramatic percent increase in Chinese consumer spending. This looks good but Figure 25.13 shows it is still pretty small compared to other countries and is a small percentage of GDP.

The governments of the US and UK have a lot of rhetoric about putting in rules to limit executive compensation and base that on the risk taken and long term performance. This sounds commendable especially if the incentive systems are changed to not encourage rogue trading. Will they accomplish anything? I am skeptical but we will see.

The G-20 claim they will curb bank pay and align economic policies so that the boom and bust cycles will not occur. They want to avoid multi-year guaranteed bonuses, have a significant portions of variable compensation deferred, paid in stock and subject to claw back if earnings decline. Banks are also encouraged to increase the quality and quantity of capital in reserve against future losses. The growing influence of China, Brazil and other emerging economies made G-20 not G-8 the key economic work group.

Fig. 25.13 China's consumption still small relative to other countries and small as a percent of GDP. Batson (2009)

The US economy is built on the concept that you get paid what you can get. And that depends on demand for services. The average late night crime or medical drama in the US costs $3 million to produce each week. Charlie Sheen basically playing himself gets $825,000 per week on the half hour sitcom Two and A Half Men for a 23-week season, some $20+ million just for him. Indeed, before his firing, he got up to a reported $1.7 million per show. He got fired for criticizing the show's producers. Now they give a much more mediocre replacement some $700,000 per show and it has dropped from #1 comedy to #3. Despite his faults, Charlie Sheen is a great actor! Many other movie and sports stars earn this much or more. As a joint Red Sox Yankee fan, I note that C. C. Sabathia, a great August (essentially all wins then) pitcher for the New York Yankees but so-so in the spring and post season (so far) has a $161 million contract. I suppose increasing the marginal tax rate on high earners could grab back some of this excess which like states and governments desperately need. But studies are mixed on the value of doing this as its a drain on spending. Some of the highly paid actors and sports stars actually do a lot of good in giving back. High on this list are Brad Pitt rebuilding New Orleans housing and Cher giving safety helmets to soldiers in Iraq and Afghanistan during the Bush administration. Whoops .. why isn't it the government paying for this one?

In our investment world we have a lot of highly paid people and it is all based on skill in managing large sums of capital. Should or will this change? It sure does not look like it.

But there remains a huge problem of six unemployed people for each new job (14.5 versus 24 million) in the US and the about 9.7% unemployment rate. The 59% recovery in the DJIA from the March 9 low has not helped these people at all. There is too much uncertainty so hiring is not happening. In fact, the government bailout went more to the Goldman Sachs and Warren Buffett types around the world. Since I work on the intellectual and practical issues of such trading, etc strategies, I am well aware of this imbalance. Crime and other social problems are likely to increase.

Figure 25.14 shows that Japan's population is dropping and aging. For more on this see our new Wiley book of Optimizing the aging, pension and retirement dilemma, Bertocchi et al. (2010).

Their debt is more than twice the US debt and rising as a percent of GDP, see Table 25.1.

Japan has been crippled by high consumer costs, substandard profit per manhour is 30% less than the US. The government has historically been focussed on protecting losers rather than promoting winners. Their growth rate is low, well under 2%. A plus is that Japan's debt is offset by the postal savings which have low interest rates. Laing (2009)argues that the JGB bond market may be the biggest bubble in world economic history. In my lectures, I have frequently said that the Japanese are the best at losing money with some notable examples. A default is possible and increasingly likely says Harvard professor William Overhold, who argues that Japan's debt is destined to keep rising (as shown in Table 25.1) because of unfunded pension liabilities for the aging population and contingent liabilities that the central government bears as a result of functionally insolvent local government.

To conclude, the most likely bubbles are in long bonds in the US, UK, Japan and elsewhere. Inflation looks ready to loom its ugly head sometime in the future but as of March 2012 is not here yet. The Fed thinks it is ready for this. We will see. Meanwhile, cash returns nothing, bonds are very risky, real estate is problematic but rising slowly from the extreme lows and stocks look like the only game in town so they likely will continue rising.

²He was bought as a yearling for $4 million - a hefty sum. But after he won the Derby, Coolmore, the wisest business people in racing, bought him for $60–70 million. Have they lost money? Are you kidding! Before it was ever found out if he was any good as a stallion, several years of more than 200 breeds at $200,000 each were collected from stud duties in Kentucky and the southern hemisphere.