Sucker for a story, or, Why you should never buy an IPO

Anyone who has attended one of my behavioural finance speeches will probably have heard me talk about the dangers of stories (Chapter 15 of Behavioural Investing) , and use IPOs as an example. I point out that IPOs have usually got wonderful stories attachted to them, but statistically they are a nightmare investment. In chapter 7 of Behavioural Finance, I present a mass of evidence that shows on average IPOs underperform the market by around 30% over the three years after their listing (after their first day of issue).

Generally during the speech I also talk about need to reverse engineer valuation to avoid anchoring on market prices. That is to say, take market prices and back out what they imply for growth, and then assess whether there is any likelihood of that expected growth being delivered (Chapter 2 of Behavioural Investing).

A new paper by Cogliati et al (available here http://papers.ssrn.com/sol3/papers.cfm?abstract_id=965450) combines these two observations. They examine some 168 IPOs that took place in the UK, France, Italy and Germany between 1995 -2001. Because of their methodology Cogliati et al end up with a sample that has actually larger and older IPOs than the average seen across the exchanges that they covered. So their results don't just reflect the explosion of small cap tech stocks seen in the dot.com boom. Additionally, as a result of this the results uncovered are likely to understate the true picture!

Cogliati et al reverse engineer the growth rate in free cash flow implied by the IPO listing price. They use a two stage DCF model with a five year abnormal growth period, before growth settles down to GDP. On average, the IPOs in their sample show an implied growth rate of nearly 20% p.a. !

The chart at the start of this entry shows the implied free cash flow growth rate for years 1,3 and 5 against the actual delivered growth rate. To say that IPOs fall enormously short of investor's expectations would be a gargantuan understatement. Far from delivering 20% growth in FCF in year one, most IPOs continue to have negative FCF. Even by year 5, they are showing a paltry 3.7% FCF growth, a far cry from the 20% expected at listing!

Cogliati et al use the realised cash flows to estimate a true 'fair value' for the IPOs in their sample. They find that the listing price is on average 70% above their intrinsic value calculations!

Focus on the facts, not the stories!

Earnings manipulation as source of short ideas

One of the wonderful things about a change of jobs is that it provides a good opportunity to clear out your old stuff. Whilst I was doing this the other day I came across a note I wrote at the end of October 2003, which I called Earnings Junkies. I scanned the note and came across a list of stocks that I had put together which had poor M scores.

The M score was created by Professor Messod Beneish. In many ways it is similar to the Altman Z score, but optimised to detect earnings manipulation rather than bankruptcy. The orginal full paper can be found here www.bauer.uh.edu/~swhisenant/beneish%20earnings%20mgmt%20score.pdf

Beneish used all the companies in the Compustat database between 1982-1992. The M score is based on a combination of the following eight different indices:

DSRI = days' sales in receivables index (measured as the ratio of days' sales in receivables in year t to year t-1). A large increase in DSR could be indicative of revenue inflation.

GMI = gross margin index (measured as the ratio of gross margin in year t-1 to gross margin in year t). Gross margin has deteriorated when this index is above 1. A firm with poorer prospects is more likely to manipulate earnings.

AQI = asset quality index (asset quality is measured as the ratio of noncurrent asses other than plant, property and equipment to total assets). AQI is the ratio of asset quality in year to year t-1.

SGI = sales growth index (ratio of sales in year t to sales in year t-1). Sales growth is not itself a measure of manipulation. However, growth companies are likely to find themselves under pressure to manipulate in order to keep up appearances.

DEPI = depreciation index (measured as the ratio of the rate of depreciation in year t-1 to the corresponding rate in year t). A DEPI>1 indicates that assets are being depreciated at a slower rate. This suggests that the firm might be revising useful asset life assumptions upwards, or adopting a new method that is income friendly.

SGAI = sales, general and administrative expenses index (the ratio of SGA expenses in year t relative to year t -1).

LVGI = leverage index (the ratio of total debt to total assets in year t relative to yeat t-1). An LVGI >1 indicates an increase in leverage

TATA - total accruals to total assets (total accruals calculated as the change in working capital accounts other than cash less depreciation).

These eight variables are then weighted together according to the following:

M = -4.84+0.92*DSRI+0.528*GMI+0.404*AQI+0.892*SGI+0.115*DEPI-0.172*SGAI+4.679*TATA-0.327*LVGI

A score greater than -2.22 indicates a strong likelihood of a firm being a manipulator. In his out of sample tests, Beneish found that he could correctly identify 76% of manipulators, whilst only incorrectly identifying 17.5% of non-manipulators.

When I ran M score on the non-financial companies in the S&P500 at the end of October 2003, some 57 stocks showed a M score greater than this critical threshold. I have recently updated their performance. Since October 2003, the S&P500 is up some 38%. An equally weighted basket of the stocks identified as potential manipulators is up only 21%, a 16% underperformance.

The median return on the identified stocks is just 12% (a 26% underpeformance of the S&P500). Some 73% of the stocks uncovered as potential manipulators had returns below the market, 20% had negative absolute returns!

As I was thinking about this issue a new paper by Prof. Beneish landed in my inbox (
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=100684).
In this new paper Beneish explores the use of the M score as a stock selection technique. Beneish examines portfolio deciles based around his M score over the period 1993-2003 with annual rebalancing done four months after the financial year end.

The results are impressive. When using market and size adjusted returns the M score strategy generates a hedged return of nearly 14% p.a. Using the Fama and French 3 factor model (market, size and style adjusted) , the stocks with the worst M scores show a -12% return, whilst the stocks with the best M scores show a 4% return, generating a long/short return of 16% (of course).

In the last year I have had many conversations with traditional long only fund managers who were setting up internal hedge funds or 130/30 style products. They were often surprised when I suggested that the shorts were not just the opposite of their longs. Shorting requires a different discipline. Perhaps using the M score might offer up a short list of potential candidates.

With earnings at a cyclical peak, finding out who has been fudging their numbers could be a particularly useful pursuit. Sadly, I don't have the data to run this screen from here, but perhaps a call to your favourite quant analyst might be in order (if that happens to be Andy Lapthorne, then you are out of luck as I know he, like me, is enjoying his garden at the moment).

The Quant meltdown or a tale of too much leverage?

As markets have declined, I've witnessed a surprising degree of schadenfreude from many commentators (including some of my friends) over the problems suffered by quant funds of late. Bloomberg has been rife with stories of problems at funds such as Goldman's Global Equity Opportunity, AQR and Renaissance.

Some have seized on these problems as 'proof' of the failure of quant investing. This is, of course, utter nonsense. No model works all the time, especially one based on valuation. For those familiar with Joel Greenblatt's Little Book that beats the market, he describes with glee the fact that the strategy fails in four years out of ten. Why? Because it means that those focused on the short term will be unable to follow the strategy in a disciplined fashion. This furthers the opportunity for willing to invest for the longer-term.

We all know that value strategies can have poor years (just recall the problems that value managers suffered in the dot.com bubble). Quant investing is just a method of removing the human from (much of) the decision making process. In the past I've written on the enormous evidence to suggest that quant models are superiour to human judgement in a wide range of fields (Chapter 22 of Behavioural Investing). The recent performance of quant funds does nothing to alter this.

However, this is not to excuse the quant funds. I've been surprised by the scale of the problem they have encountered. When I looked at the performance of a generic value strategy (such as price to book), the last few months have not been pretty. For instance, in the US a strategy of buying the cheapest stocks by P/B and selling the most expensive has generated a 3% negative return in July (and presumably worse in August - although I don't have the data to prove that).

Now a -3% return isn't that bad. It actually isn't anything out of ordinary for such a value strategy. The chart at the start of this entry shows the distribution of returns to the price to book strategy since 1926 (thanks to Ken French for the use of his data). In fact returns of -3% or more per month have been seen 10% of the time.

The mean return of the price to book strategy in the US is 0.4% p.m with a standard deviation of 3.5%. If we assume that the returns to value are normally distributed (obviously not a great assumption) then 95% of monthly returns should lie between 7.5% and -6.7%. So even a cursory glance at the data would reveal that July is nothing out of the ordinary.

How did the quant funds manage to translate this relatively normal occurrence into something so performance destroying?

I can think of three possible routes (not mutually exclusive) :

1. They hadn't checked the long term data.
   
2. They were using leverage.
   
3. The trades were overcrowded
   

The first path is a major problem for many risk models that use short time periods to estimate their parameters (Chapter 36 of Behavioural Investing has more on this). It should be less of an issue in this context because we last saw a value meltdown in 2000 (i.e. relatively recently), so many managers will have already been aware of this.

The second route involves the use of leverage. Value strategies and leverage don't make easy bedfellows. Indeed John Maynard Keynes observed the dangers of combining a long-term view (which any value strategy must be - Chapter 31 of Behavioural Investing) with leverage. He opined "An investor who proposes to ignore near-term market fluctuations needs greater resources for safety and mist not operate on so large a scale, if at all, with borrowed money."

When I first heard that some of the quant funds were in trouble my first reaction was that they had been deploying leverage. This has recently been confirmed by a conference call between US analysts and the CFO of Goldman Sachs (taken from Street Events).

Susan Katzke, Credit Suisse:

Okay. And just I don't know if you covered this with Roger -- I might have missed it -- but in terms of the leverage in the funds, what are the leverage parameters, and where do you expect them to be going forward? Were they in retrospect a little bit higher than you would have liked them to have been or will be going forward?

David Viniar, Goldman Sachs CFO:

They were higher than we wished they were given how fast the market moved, but they were right in line with what had been expected. And the leverage at GEO as we sit here now is around 3.5 times, which is actually a little bit under where we had told people we would operate but probably around where we will operate going forward.

Susan Katzke, Credit Suisse:

Okay. And the 3.5 times just to clarify is with the $3 billion equity investment?

David Viniar, Goldman Sachs CFO:

With the equity investment.

Susan Katzke, Credit Suisse:

Okay. So closer to 6 times before that.

David Viniar, Goldman Sachs CFO:

That is correct.


So Goldman's Global Equity Opportunity fund was running with leverage of 6 times! This strikes me as madness for a value orientated strategy.

Overleveraged trades often go hand in hand with over-crowded trades. Cliff Asness at AQR wrote the following to his investors "This isn’t about models, this is about a strategy getting too crowded, as other successful strategies both quantitative and non-quantitative have gotten many times in the past, and then suffering when too many try to get out the same door. We knew this was a risk-factor but, like most others, in hindsight, we underestimated the magnitude and the speed with which danger could strike."

If there is something to be learnt from the problems the quant fund find themselves in, it is far less to do with the failure of quant and far more to do with the dangers of leverage.

Fear and greed

In one of my last notes for Dresdner Kleinwort I stated that perhaps I had lost the plot. When I looked at markets all I could see was irrational complacency, the pricing of assets without any regard to risk, and the dash to trash.

Emerging markets were trading at the same valuation multiple as developed markets. Small caps were on a premium to large caps (ie a negative liquidity premium), and junk equity was the flavour of the day.

Given the unwind we have witnessed in the markets, it is fair to ask how far along are we in the process of risk re-pricing?

The chart above shows my fear and greed index. A simple construct that measures the risk adjusted relative performance of global equities vs global bonds. Whilst it has fallen sharply, it remains a long way from the kinds of panic sentiment that usually characterize buying opportunities.