Quantifying impact of trading costs

Academics in their studies have come up with various estimates of the costs of trading to individual investors:

·         Meyer  estimates at 0.90% the typical cost (excluding taxes) of trading for Germany equities investors; see PDF doc.2220

·         In 2000 Odean and Barber, for each round trip by discount broker clients in the 1990’s, assumed that the average trade cost was about 1% in spread costs and 3% in commissions (again excluding taxes); see Odean and Barber 2000, p.775, 780  PDF doc.2209D.  

Almost none of the estimates even try to measure market impact or capital gain acceleration costs. The estimates of commissions and spreads seem much higher than those currently in Canada, at least for customers of discount brokerage firms trading online in modest amounts of large –cap, widely traded stocks; see RetailInvestor.org . The estimates of commissions may not be far off the mark for customers of full-service Canadian brokers.

By comparison, the cost of trading by actively-managed mutual funds has been more closely looked at, but as an annual cost (not the cost per trade):

·         In a commentary on mutual funds, after consulting different sources,  we used as an estimate of the impact of average annual turnover cost on returns of the typical actively managed equity mutual fund (excluding taxes) to be equal to 1.28%; see How much do actively managed mutual funds cost .

·         In the same commentary we estimated that the impact on returns of fund investors due to acceleration of capital gains tax from excessive trading by the typical actively managed equity mutual fund to be in the order of 0.89% per year.

Quantifying trading impact on gross returns

The impact of trading on gross returns (i.e. before transaction costs) can refer to actual losses resulting from unprofitable trades, or more commonly refer to an estimate of by how much the returns of a portfolio are less than (or sometimes exceed) a benchmark, passive portfolio or index or to the returns of investors less inclined to trade. The return deficit is generally blamed on poor securities selection and/or market timing skills. If a lack of skills is found, it would be expected that the more one trades, the greater the deficit.

Active trading by retail investors is often based on their perception of where individual stocks are going (stock picking) and/or where markets as a whole are headed (market timing). Investors develop or adopt rules of thumb to attempt to predict movements in stocks or markets, but they are largely useless; see Vanguard Forecasting stock returns or as a PDF doc.2230 . Various studies have been made to try to compute the impact of trading on financial returns. The impact is typically negative both on a gross basis (i.e. even without taking into account trading costs).

Asset allocation is the main determinant of the return of a portfolio. Rebalancing of a portfolio from time to time to bring it back in line with the investor’s target allocation is a commonly recommended technique. Regrettably retail investor trading often greatly exceeds the degree of rebalancing in commonly used approaches; see portfolio rebalancing implementation: the nuts and bolts . Even worse, such trading often tends to be the reverse of proper rebalancing. In other words, in proper rebalancing an investor sells an asset class that has risen in value and reinvests in an underperforming class, while most retail investor trading tends to be the reverse, selling losers and buying winners.

Studies of individual investor trading have tried to estimate the adverse impact of excessive trading and other poor investor practices; Odean and Barber in 2011 doc.2209F, Meyer at al doc.2220; and Carpentier and Suret doc.2225 in 2012 surveyed the recent literature. The estimates trading impact are all over the map, and have been criticized; see RetailInvestor.org .

Non-USA market studies

We begin with studies of 3 markets outside the USA:

·         Steffen Meyer and 6 others from Goethe University carried out a review PDF doc.2220 of the recent literature on returns of individual investors. They then present a study of trading by German investors at an online German discount broker between 2005 and 2010. They argue (p.3) that online investor are more sophisticated than the average investor, and therefore expect any results to be slightly overstating the skill of investors if there is any bias. They estimate that what they (charitably) call insufficient skill results in an adverse impact on gross returns and on  returns net of expenses and trading costs (which they estimate at 90 basis points per year per investor on average- p.14). They state that the  results are a clear case for passive strategies (p.30):

• Analyzing gross returns, we found that about 89% of individual investors have negative skill. This implies that 89% of investors underperform the market when pursuing an active strategy – unless luck is on their side. The magnitude of the average skill of all investors is also very large with about -7.5% per year. Taking expenses and trading fees into account, the figures are even worse: 91% of individual investors have negative skill which does not even suffice to cover their expenses and trading fees. As expenses and trading fees amount to about 1% per year, the average skill of all individual investors amounts to -8.5% per year. It can thus be concluded that the large majority of individual investors do not have skill to outperform the market – and if they do, it is mere luck. p. 29

·       In 2008 Bauer et al. PDF doc.2211 compared option and equity trading retail clients of a large Dutch discount broker. As an initial observation they found that online investors trade 3X more than offline clients i.e. clients of full-service brokers. They found that investors at the discount firm who traded options had gross returns 1% per month less than the returns of equity-only traders at the same firm, and the difference after costs increased to 2.75%. They attribute the poor results of option trading first to poor market timing, and the results are made worse by transaction costs.

·   I    In 2009 Odean, Barber and 2 Taiwanese academics PDF doc.2209B looked at the transaction data for trading on the Taiwanese stock market by all investors between 1995 and 1999. Trading commissions are typically around 0.14% to which is added a transaction tax of 0,3%.p.612. Total annual stock turnover on the exchange is around 300%, triple that on the NYSE. P.613.Their simple conclusions are that Individual investors pay an exorbitant price for trading actively. Individual investors could participate in financial markets at low cost by following a simple buy-and-hold strategy.p. 628. They blame overconfidence and the desire to gamble account for much of the active trading and substantial losses of individual investors in Taiwan.p.628.

H    Here are their more detailed observations:

Our empirical analysis presents a clear portrait of who benefits from trade: individuals lose, institutions win. P.610……Put differently, it is a 3.8 percentage point annual reduction in the return on the aggregate portfolio of individual investors. These losses can be broken down into four categories: trading losses (27%), commissions (32%), transaction taxes (34%), and market-timing losses (7%). The trading and market-timing losses of individual investors represent gains for institutional investors. The institutional gains are eroded, but not eliminated by the commissions and transaction taxes that they pay. P.611…. Stocks sold by individuals outperform those bought.p.618,620… This observation is quite consistent with models  that assume investors are overconfident and, as a result, trade too aggressively and to their detriment. P.622.

USA

Strangely enough the studies of individual US investors are less exact. Odean and Barber are the most prolific US writers on this subject; see:

·         Trading is hazardous to your wealth 2000 doc.2209D 

·         Why do investors trade so much 1999 doc.2209C  and 2000 (DFA sponsored) doc.2209H

·         Boys will be boys 2001 doc.2209A

·         Behavior of individual investors 2011 doc.2209F

·         Systemic Noise 2006 doc.2209E; and

·         Impact of attention and news on trading 2008 doc.2209G.

In 2000 Odean and Bartber doc.2209D p.776, using similar data from 1991 to 1997, found that it is the costs and frequency of trading alone that explain poor investment performance compared to the market as a whole; the 2000 study looked at the aggregate performance of all stocks held by the investors, not just those traded- p.777.

In 2000 they also found that high trading investors at US discount brokerage firms underperformed both low-trading investors and market indexes by 7.1% and 1.5%, respectively; they attributed the underperformance, however, entirely to transaction costs i.e. they found no difference in gross returns-p.774.

In a 2000 DFA sponsored text doc.2209H they estimated at 5.5% the (negative) difference in returns between active and low-trading US discount broker clients.

 In their 2008 study PDF doc.2209B of the Taiwan market they stated:

Less comprehensive studies suggest that trading losses and costs for individual investors in the United States are about 2 percentage points a year (Barber and Odean, 2000, 2001). (U.S. individual investors trade less actively, but run a higher risk of trading with better-informed institutional investors.)

And in a 2011 paper doc.2209F p.4, 15 they state that transaction costs are not the whole story. Individual investors also seem to lose money on their trades before costs.

It is clear that Odean and Barber consider trading to be detrimental. However, what quantitative estimate of the impact of trading on gross returns of US investors should be drawn from their studies as a whole is not entirely clear.