Buying, holding, & rebalancing outperforms prediction shifts & market timing

By Keith Matthews

 

The last six months or so have seen the S&P 500 and other markets soar to historic heights. These never before seen market highs have led many to question whether we are flying too close to the sun, and if we should brace ourselves for a correction. Open up a newspaper or go online and you will find scores of articles and think pieces claiming to tell you what the market will do and what you should do about it. Yet which of these seemingly all-knowing (yet all too often contradictory) forecasts should you believe? What decisions should be made for your investments today, tomorrow, or six months from now? Should we even be doing anything at all?

At the beginning of 2016, the Royal Bank of Scotland advised investors to “sell everything” ahead of what they claimed would be a disastrous year for stocks. The sky did not fall, and a year later the markets achieved record highs. What went wrong for RBS? It turns out that predicting the future of any complex system is hard work; doing so consistently is impossible. The track records of many notable ‘expert’ forecasters leave much to be desired.

Empowering our clients on this subject matter is important to us. Helping our clients tune out the noise not only produces better investment results; it allows us to focus on matters that we can control (long term asset allocation, cash flow spending, and tax efficiencies). In addition to better long-term returns, tuning out the noise also reduces stress along the investment journey. We are firmly committed to this value proposition.

The problems with predictions

Problem #1:

Predictions are so pernicious because of how believable they appear on the surface. Market predictions are typically made by so called (or self-styled) industry experts who are typically perceived as credible sources of information. Their predictions are usually part of a larger narrative, which plays into the notion that we can identify how events in the story will play out before they do. These scenarios are seemingly backed by data and research, but oftentimes they ignore information that would contradict the prediction, or instead provide a limited/ incomplete range of information.

Problem #2:

Market predictions often motivate investors to make changes to their long-term asset allocation mixes. This is called market timing. No one (including market gurus and industry experts) can accurately predict market movements on a consistent basis. There are simply too many moving parts in any economy, let alone in today’s global economy, to accurately predict the different factors that drive asset class prices up or down.

Experts may get a market call right once or even twice in a row, but to add value long-term they need to be consistently right. In other words, they can’t afford to make a mistake. The evidence shows that it is incredibly unlikely that any market expert or firm can consistently predict the future over the course of your investment horizon, be it 10, 20, or 30 years and more.

Who is making the predictions, and why are they so confident?

It is important to understand and accept that the financial industry depends on predictions to build its value propositions. Wall Street, Bay Street, the brokerage industry, and the financial advisory business have used slick marketing campaigns to position themselves as experts at predicting the markets.

Stock market forecasters are incredibly confident when conveying their message about the market’s near-term direction. Whether they appear in an interview or write research notes – their confidence in their predictions is remarkable. However, a look at the data shows that these market pundits have a poor batting average. How could this be?

David Dunning and other researchers have documented a troubling pattern. They discovered that individuals who rank among the worst predictors tend to be the ones who have the most confidence in their abilities.

To test this troubling pattern, Dunning and his colleague Justin Kruger asked people to rate how they performed on a logic and reasoning test. Figure 1 splits the different ratings for perceived ability, perceived score, and actual score into the four quartiles based on performance. Remarkably, those who did not perform well still had very high levels of confidence in their perceived abilities.

Unskilled and unaware of it

 

 

 

 

 

 

 

 

It’s no wonder then that market experts are all so confident in their predictions.

Predictions in 2017

Many current predictions point to the elevated asset prices (stocks and bonds) and how this will play out badly for investors. The truth of the matter is that no one really knows. There is no arguing about “if”. What cannot be predicted is when, how deep and how long lasting the drop will be, as well as the start and strength of the recovery. To maximize your profit, one would need to accurately predict these five variables.

The table below shows some basic market statistics from the six previous market corrections. It is remarkable how the variables are distinctly different with each correction.

Major corrections in the past 50 years

For a 65% equity, 35% fixed income portfolio*

 

 

 

 

 

  1. % drop from the previous market high to the market trough (low)
  2. Number of months of market decline to market trough
  3. Number of months of market ascent to recover to previous market high
  4. Total number of months to go through the full cycle
  5. 5 year cumulative gain from that market low

*Equity divided equally between: TSX Composite Index, S&P 500 Index and MSCI EAFE Index (net div)

There is no refuting the fact that the market is more expensive today than it was five years ago. Yet is it possible for us to consistently profit from this type of information? In a remarkable study on this very question, the London Business School team of Elroy Dimson, Paul Marsh and Mike Staunton looked at the price-dividend ratio (cyclically-adjusted) of the inverse of the dividend yield.

The professors used a market-timing model based on information that investors would have known at the time, dating all the way back to 1920. The model required investors to sell equities to cash every time future real returns looked low (or stocks were expensive) and buy every time real returns looked high (or stocks were cheap). Sadly, this model was a complete flop; in every single market (out of 20), the approach generated a lower return than investors would have achieved had they just used a buy-and-hold strategy and stuck with equities.

There is no easy way to time the market; otherwise it would have been discovered, exploited, and eliminated long ago. However, what can be concluded is that when asset class prices are high, expected future returns will be lower. We will discuss the concept of lower expected returns further in a future article.

How to act now

Our investment philosophy guides us when markets are high just as well as it does when markets are low. By adhering to our philosophy and core principles, we are better equipped to handle the shocks that every investor encounters over the course of their lives. Money is an emotional topic, but it is imperative that investing remains a rational endeavour. Instead of dwelling on short-term fluctuations, we will continue to manage your portfolio to help you reach your long-term goals.

There have been corrections for as long as there have been markets, just as there have been crashes and recoveries. Our client portfolios are built to operate effectively no matter what is happening on the world stage or on the stock market. We will not make predictions about how or when the next correction will occur, just as we would not recommend our clients take further action with their portfolios. It can be tempting to try to time the market or use tactical asset allocation, but the data show time and again that staying invested and rebalancing to your investment policy is the best approach for your long-term success.

_______________________________________________________________________________________

 

Special thanks to Connor McRae (Associate at Tulett, Matthews & Associates), for his research and support in this article