Bears to sell in May and go away

The latest reason for the bears to sell: it’s May — “Sell in May and go away” doesn’t work well in Asia ex-Japan. May to July have on average been positive return months with only August and September being negative. If you believe that we are in year three of a five-to-six-year Asian bull, then selling in May is the wrong thing to do.

Investors retain sizeable walls of worry and sentiment indicators corroborate this. The time for concern about equity markets is when investors no longer worry and euphoria prevails. We have some ways to go until then.

Not all countries and sectors perform alike — the weakest returns over the summer months have been in Malaysia and the Philippines whilst the best returns have been in India and Indonesia. For sectors, technology has historically done well whilst banks have not enjoyed the summer heat. In terms of earnings revisions, March, May and September have historically been the weakest.

Our own sentiment indicators and conversations with clients reveal the same thing: There isn’t a great deal of optimism around these days, especially when it comes to equity markets. Most investors have a long list of concerns and a list of major events. Aside from the usual — inflation is out of control, China blows up, QE2 ends — we have US debt default, European debt/default risk, and developed markets (DM) vs emerging markets (EM).

Given that many risk events YTD (DM vs EM, inflation fears, rate fears, oil price increases, margin pressure and the Japanese earthquake) have failed to lead markets lower, the next concern aside from the end of QE2, is “sell in May and go away”. More on this in a second.

What we find fascinating about the current market cycle is that if one had sat down in 2009 and said, “I’ve no idea what the recovery will look like, but there will be one so I’ll order the usual please” that investor would have been pretty nearly spot on. For anyone who thought it would be a different recovery and there were clear reasons to think it would be, all his thinking would have added up to little more than underperforming the index.

In the same vein, valuations today of 14.6 times trailing P/E, compared with historical average P/Es of 15.1 times at the end of year two and then 15.5x at the end of year three. Again this is more similar than dissimilar. The same is true when it comes to which investment factors are adding to returns and which are not. In prior cycles, come year three, a combination of price momentum and earnings revisions have proven to be particularly rewarding. Guess what; it is so again. Risk as a factor is underperforming, indicating that not all hands are on deck and committed.

We still have a wall of worry. It is hard — given historical evidence — to make the case that the world is a riskier place, or that it has more moving parts.
We now have second-by-second updates, but this does mean more moving parts. Last month, Asian equities were to be sold due to higher oil prices; today the price of oil is correcting and markets sell off. Cake, eat it and then still have it come to mind.

The wall of worry is there to be climbed and we expect markets to climb it for the remainder of the year. Equities are not expensive on an absolute basis, relative to their history vs bonds or cash and even vs physical assets like real estate in Asia ex-Japan. Don’t forget, Asian earnings have surpassed their prior peaks, equity markets have not, nor have valuations.

Sell in May and go away …maybe better to sit out August and September
Figure 1 looks at monthly returns for Asia ex-Japan over the past 30 years. On average, three months have given investors poor returns. These have been March, August and September. Again, on average, May has not been such a bad month, up 1%, June was a little worse at +0.4% and July came in at +1.3%, the highest returns out of the three-month period. Sell in May and go away really only works due to poor performance in August and September. For all the talk of how bad October is, on average, it has actually been an up month, rising 1.1%.

On average, all months are up months. April, December, October and July are the highest-return months. Sell in May and go away proves to be an abject disaster that costs investors dearly.

The call here is not so much “sell in May and go away”, but whether you believe that we are in year three of a five-year-plus up cycle or in a cycle that is about to roll over. We still see no evidence to suggest that we are in for a short cycle with a recession around the corner. Market performance, valuations, and factor behaviour indicate that this time will be no different than prior cycles, which have been five to six years in duration and end as P/BV approached/reached three times book.

Not all markets or sectors react the same. In May, best avoid Thailand and South Korea as both show negative returns, and go long Indonesia and the Philippines. In June, stay in Indonesia but also pay China a visit, but avoid the Philippines and Malaysia this time. You get the drift. Over the whole summer period, historically the best returns have been in India and Indonesia. Malaysia and the Philippines have a tendency to show the most negative months.

In terms of sectors, the month of May has little to offer investors in technology or other financials. On average, they do poorly. However, those long energy and real estate can rejoice at super normal returns. June is still a good month for energy and is joined by technology (mean reversion from the weak prior month). Those wishing to enjoy a quiet time on the beach over the summer should avoid banks as they do poorly but be long technology, which has on average done well during this time.

Usual to have a correction in year three of the cycle
It is normal for market corrections during the 3rd year of the equity cycle. By year three, interest rates are usually on an uptrend as monetary policy goes from loose to tight.

This is a cause of uncertainty. Corporates begin to increase capex, which near term is both a negative for ROE and a drain on over all liquidity. So as per Figure 2, we have seen many corrections in Asia ex-Japan. These occurred on average at a P/BV multiple of 2.3x and ended on a P/BV multiple of 1.7x. Based on current valuations, this would imply that the MSCI Asia ex-Japan index moves back to its 200-week moving average of 475 vs an index level of 580 today. A slightly aggressive call but it’s happened before.

Last year it took the MXASJ three attempts to breach the 500-index-point level. It broke through on the third try. This year the MXASJ has had one attempt at the 600-level followed by a small sell-off then another retest of 600 followed by a sell-off… Will the third time be lucky again? At the time of writing, both the 50-week and 200-week moving averages are still moving higher, suggesting that it is still premature to make the bear call.

Fund flows are also seasonal by nature
Since 1992 — the 1st date for fund flows — there has been a distinct pattern to foreign flows.

As expected, this has a knock-on effect on the market. 1Q flows tend to be positive, peak in 2Q and decelerate into 3Q before picking up into 4Q. Flows equal demand, so less demand for equities means prices will clearly be softer. Year to date, things have been slightly different: money has flowed out of Asia ex-Japan rather than into Asia ex-Japan. So much for a strong 1Q.

Earnings revisions weakest in March and September
We know that earnings revisions work in terms of helping generate excess returns.

In Asia ex-Japan, the two worst months for earnings revisions have historically been March (just before the results…information leakage?) and then again in September when it becomes obvious that the full-year numbers won’t be met. So it should come as no surprise that September is the weakest month on average, not the period before.

Seasonality in markets also means seasonality in factor returns
In the same way that we have seasonal fruit and vegetables, so do certain investment factors do better than others over time.

Growth investing over 1995-2010 has historically done well in the early part of the year, only to see its efficacy diminish as we move into the later quarters of the year.

Against that we have momentum investment styles. After a rather poor start to the year, it is a case of buy in May until June and return to momentum as an investment style in September. YTD, it would seem that performance has been better than average.

Finally, we have value — the consistent one. Whereas both growth and momentum investment styles have some months of negative average month-on-month returns, this is not the case when it comes to value as an investment style. Value may be less exciting than growth or momentum investing but it surely reduces the blood pressure. - Written by Citigroup   



This article appeared in The Edge Financial Daily, May 10, 2011.