Behavioural Traps for Active Managers to Avoid (Part I)

Behavioural finance is the intersection of applied psychology and the behaviour of financial market participants.  The likes of Richard Thaler, Nicholas Barberis, James Montier, Nobel prize winners, Robert Shiller and Daniel Kahneman, and many others have done a tremendous job in elucidating the heuristics or mental rules of thumb that investors (and broking analysts) rely on, which contribute to systematic errors in judgement and decision-making.

Some of these heuristics include: representativeness (leads to extrapolation), availability (leads people to over-estimate the probability of familiar and salient events), anchoring or framing (can distort decision making by affecting a person's frame of reference) and self attribution or over-confidence (people attribute their good performance to skill and judgement, and poor performance to bad luck or factors beyond their control).

Although these errors are not unique to financial market participants, it is reasonable to think that such errors can be costly, particularly when portfolio managers are managing billions of dollars.  If enough investors suffer from such biases they also lead to significant limits to arbitrage, which imposes costs and risks on investors who are less ‘irrational.’

2014 has proven to be another difficult year for active fund managers in Australia and worldwide for cyclical and structural reasons.  The unexpected decline in sovereign bond yields confirms that investors eschewed risk from their portfolios.  Further, the structural shift towards low cost beta like products that commenced since the financial crisis continued, as investors remain sceptical of the ability for active funds to consistently deliver alpha after costs and fees.

At a time when the philosophy of active management is coming under attack as active funds continue to lose market share to ‘dumb’ and ‘smart’ beta products, it behoves PMs to be cognisant of, and avoid behavioural traps.

Hindsight bias

Hindsight bias is better known colloquially to many people (at least in the United States) as the ‘Monday quarterback’ phenomenon.  In the best seller, Expert Political Judgement, Phillip Tetlock argues that hindsight bias amounts to discounting counterfactual scenarios that might have realistically emerged when considered in real time.  He provides compelling evidence that political scientists are prone to look back and believe incorrectly that they had predicted certain outcomes that came to pass.

It is far too easy for fund managers to also fall into this trap because they forget what their views might have been in real time.  After all, financial markets are noisy at the best of times, particularly as the signal to noise ratio is typically low.

There is good reason why financial market participants suffer from hindsight bias.  In an industry where the ability to articulate a convincing story is just as important as track record to growing assets, over-confidence can be a valuable commodity.  But hindsight bias and over-confidence are detrimental to delivering alpha because they limit the ability for broking analysts and investors to learn from past mistakes.

The obvious way for PMs to avoid this behavioural trap is to diligently document their views in real time, which provides scope to look back and benchmark those views with events that subsequently transpired. 

The process can also help to disentangle the role of luck from skill.  For instance, an overweight position in Qantas in early 2014 might have been justified on the prospect of a CEO succession event in the near term.  No such event transpired, but the overweight position has been profitable, based primarily on the precipitous fall in crude oil prices that few had predicted at the start of the year.

The stock market is not a discount store

Profit warnings are a godsend to brokers.  They provide much need volatility and generate trading opportunities largely because they garner the attention of PMs, some of whom like to engage in bargain hunting.  The problem is that the stock market is not a discount store.  The discount store mentality arises due to framing, which distorts the investor’s perspective because they have been conditioned or become accustomed to a higher stock price.  So a profit warning and a sharp fall in price triggers a view that a stock is cheap and thus offers a bargain.

In late 2013, QBE issued a profit warning, citing problems in their North and South American operations, lower than expected investment earnings stemming from the US Federal Reserve’s zero interest rate policy and the persistently high $A.  The stock subsequently fell by over 30%, representing an 8 standard deviation event.  Analysts downgraded their earnings forecasts by around 20%. 

On face value, this amounted to a market over-reaction.  But it is reasonable to think that investors were assigning a higher beta and discount rate to the company’s future growth prospects, thus accounting for the stock’s sharp price decline.  After all, this was another in a long line of profit warnings, the business is a complex and globally diversified insurance company with poor visibility in its far flung operations, and management continue to deal with the legacy of two decades of acquisition led growth.

A number of broking analysts at the time upgraded their recommendation to buy, thanks to the view that the profit warning represented a clearing of the decks of sorts, particularly as it was accompanied by the resignation of the company’s chairperson.  Some of those analysts – influenced by the sharp rise in global bond yields in May 2013 associated with the ‘taper tantrum’ – also believed that interest rates would continue to increase over the course of 2014 associated with a normalisation of US monetary policy, which would be a tailwind for the company's investment earnings.

As it happened, yields subsequently fell sharply in 2014, most of the world’s central banks maintained zero interest rate policy settings, and despite the stock bouncing in the month after the profit warning, twelve months down the track, QBE has performed broadly in line with the market (see chart).

Who’s afraid of stocks trading at a 52 week high?

Price momentum is one of the most widely studied and robust pricing anomalies documented in equities and other asset classes, across time and countries.  Stocks that perform well over six to twelve months tend to continue to yield strong returns for up to six months.

Journal research from the United States provides an interesting twist on the momentum anomaly; stocks trading at or close their 52 week high continue to yield strong returns in the short-term. 

The authors offered a behavioural explanation; many investors are put off by buying such stocks due to concerns that the horse has bolted; a corollary of the discount store framing.  An institutional explanation might also be at play.  Portfolio managers may believe that they have reputational risk to manage if they buy a stock at its peak and it subsequently under-performs, based on concerns at being seen to be swayed by investor sentiment.

My back-testing work for Australia over the past decade shows that stocks trading at or near their 52 week high have continued to outperform in the short term, and 2014 was no exception.  An equal weighted portfolio spread across the 15 large cap stocks trading at a 52 week high at the end of 2013 has yielded a total return of 6.7% year to date, well above the 4.5% from the ASX100 index.  A number of stocks underpinned the outperformance: RHC, SEK, TAH, TLS and AMC, while the key drags were NVT and TPI.

Given that investors eschewed risk in their portfolios over the course of 2014, it is little surprise that the current basket of stocks trading at or near their 52 week high has a strong defensive theme: CTX, GPT, RMD, TLS, TCL, SYD, ANN, COH, NVN, SKI, IPL, APA, ORA, CBA and RHC.

Do not be seduced by a company’s growth prospects and the art of storytelling

A cornerstone of behavioural economics and finance is that we rely on heuristics or rules of thumbs to help us deal with information overload, since humans can absorb and process only so much information.  Financial markets are an apt setting to investigate these heuristics because portfolio managers are confronted with tremendous amounts of information, much of which is noise.

Story telling iss an effective way to help people to make sense of their world, as well as process and communicate information.  People can relate to stories, they can benchmark the stories of others with stories of their own, and stories garner people's attention.  Many CEOs understand this and are able to articulate a compelling story to buy their stock based on the prospect of strong growth.

Buying a basket of stocks with high expected growth in profitability at the end of 2013 has yielded an equal weighted return of over 10% year to date, well above the index return.  But this represents the exception; these stocks have typically delivered dismal returns over the sweep of the past decade (chart is available on request).  The third quartile or mid-growth stocks represents the sweet spot.

The poor returns from high growth stocks is not unique to Australia; the high growth anomaly is robust across the United States and other countries.  The poor performance reflects a number of factors.  Seduced by the prospect of strong growth and possibly some compelling storytelling, investors are lulled into over-paying for future growth.  Second, broking analysts and PMs tend to extrapolate past growth too far into the future.  Third, investors under-estimate the level of risk and uncertainty associated with high forecast growth.

Conversely, the mid-growth stocks are perceived by many to be dull, offering the prospect of little capital growth, and pay out high dividends due to limited growth opportunities.  Because they do not provide as a compelling story as the blue sky embedded in high growth stocks, PMs are prone to under-pay for these attributes.

At present, stocks with high forecast growth (stocks to avoid) are: ILU, OSH, ALL, CSR and BSL, while mid-growth stocks (stocks to buy) include: ANN, TLS, PRY, SGM and TAH.

The illusion of control and the (limited) influence of CEOs

Individuals - particularly those that are over-confident - typically over-estimate the control they have over events, and discount the role of chance or factors beyond their control.  Since many financial market participants are over-confident, PMs need to be careful that they do not over-estimate the importance of management quality on company performance.

Many PMs like to showcase their ability to identify and only invest in companies with good quality management.  CEOs are the custodians of shareholders’ funds, so skilled managers whose interests are aligned with shareholders ought to be necessary conditions for a company to deliver sustainably strong returns.

But factors beyond the control of the CEO and management affect the profitability and performance of a stock, including the state of the domestic and global economy, outlook for interest rates, the supply and demand for credit, shifts in technology and consumer tastes, unexpected changes in the regulatory environment, health of the banking system, and investor sentiment, just to mention a few.

Sydney Airport (SYD) demonstrates that good quality management can have little bearing on a stock’s performance.  SYD is a monopoly asset that generates highly predictable cash flows, with limited exposure to the economy primarily through passenger numbers. 

Shifts in the $A are broadly neutral for passenger numbers; a fall in the currency this year encouraged a lift in overseas arrivals but reduces the number of Australians travelling abroad.  Passenger numbers in the past have been adversely affected by factors beyond the control of management, including the spread of the SARS virus and the s11 attacks.

More recently, the strong outperformance of the stock reflects factors that have little to do with management: the strong appetite for safe assets, subdued top line growth for most companies thanks to the economy remaining stuck in a nominal recession, the decline in sovereign bond yields and continued strong performance of the yield trade.

PMs should not entirely discount the role of management and the incentives they face as a guide to the alignment (or non-alignment) of interests with those of shareholders.  But they should not discount the role that systematic factors outside of a CEO's control play, in driving stock performance.

Part II

In the second instalment, Evidente will discuss other behavioural traps for active managers to avoid, including: over-estimating the precision of DCF valuations, dangers associated with neglecting the regulatory and fiscal landscapes, the problems of chasing down answers to questions that don't matter, and the importance of seeking out alternative and objective sources of information.