The ultimate combination of value and momentum strategies depends on the investors’ time horizon, with a new institutional theory rationally explaining these anomalies by investor flows. Dimitri Vayanos, professor of finance at the London School of Economics explains why this theory can add to the practical debate of what is the best asset mix for investors.
Most active investment management strategies involve the prominent market anomalies of momentum and value, and the question of how to best combine them depends ultimately on the time horizon of the investor.
Professor of finance at the London School of Economics, Dimitri Vayanos, says that momentum and value are puzzling, and hard to understand, within standard finance theory, with some studies explaining the strategies by looking at behavioural economics and some using market frictions including delegation and agency, and fund flows.
Explaining how or why these anomalies arise is important, Vayanos says, because then investors can exploit them.
Speaking to delegates at a London School of Economics roundtable, hosted by www.top1000fuds.com, Vayanos explained a flow-based explanation for return predictability, and the idea that momentum is driven by flows.
He defines momentum as the tendency of recent performance to continue in the near future, reversal as the tendency of performance over a longer history to revert, and value, closely linked to reversal, as the ratio of market price to book value of equity or earning predicting inversely future performance.
He says that understanding flows is important to understanding momentum.
Momentum, reversal and value are well-documented empirically, and form the basis for most active investment strategies but most investment strategies are data driven through backtesting and there is not a theoretical understanding of why momentum is indeed momentum. There is not an underlying framework for understanding the anomalies.
“Fund flows generate comovement. Following outflows from some funds, all assets held by the funds drop in price,” he says. “Fund flows also generate lead-lag effects. The price drop of one asset predicts that the other assets held by the same funds will drop in the short run and rise in the long-run.”
He says that momentum, reversal and comovement are larger for assets with high idiosyncratic risk
“Trading against mispricings in those assets subjects fund managers to high risk of underperforming their benchmarks,” he says. “There is evidence of associating earnings to predictability of returns – value stocks have high expected returns and low and declining earnings.”
The portfolio management implications include the question of how best to implement value and momentum, how they should be combined, and how the ultimate strategy connects with the time horizon of the investor.
He claims that his modelling can help answer those questions by looking at the response of mutual fund flows to performance and the price impact of those flows.
The results show there is a negative correlation between momentum and value, and diversification benefits from combing the two.
In terms of risk, momentum is a series of individual events so the long-run risk of momentum is a sum of short-run risks. But value is a long term game, if a stock does poorly in one year it will become even cheaper.
“The long-run risk of value is smaller than the sum of short-run risk, so value overtakes momentum for long investment horizons,” Vayanos says.
One of the delegates, Olivier Rousseau, executive director of the $50 billion Fonds de reserve pour les retraites, asked why investors should bother with value, when “you can get all the goodies from momentum over and over”.
But Paul Woolley, founder of the Paul Woolley Centre for the Study of Capital Market Dysfunctionality at the London School of Economics and a co-author of Vayanos, says that momentum is flattered because it is data driven.
“It is treacherous, it is fool’s gold,” he says.
Vayanos says that funds should invest in both momentum and value but the combination depends on the time-horizon of the investor.
“If there is a shorter horizon then a focus on momentum, a longer horizon then there should be larger weights on value,” he says.
In addition, from the point of view of a long-horizon investor, momentum’s attractiveness is further reduced because it has an anti-hedging property.
“Momentum does poorly in our framework when mis-pricings decrease which is when investment opportunities worsen. In particular poor performance of momentum predicts poor future performance of value. When momentum performs poorly, therefore, investors are not compensated by having access to other investment opportunities that are expected to perform well.”
Wilma de Groot, vice president and portfolio manager of quantitative equities at Robeco, said the presentation was excellent in understanding the drivers of momentum and value that are used in investing.
“It helps to rationalise the momentum and value effect when usually they are related to irrationality, it is a very good model,” she says.
Tony Broccardo, chief investment officer of the $30 billion Barclays UK Retirement Fund asked how the delegated relationship between asset owners and fund managers relates to the momentum and value phenomenon.
“Our theory emphasises performance-driven flows, but flows can also be for other reasons. There is empirical evidence that more delegation causes more momentum, although more study is needed. If there is more institutionalisation then other things change so it is hard to define each characteristic but we should also look also at benchmark design,” Vayanos says.