Skip to content

Short-term vs Long-Term Market Efficiency

October 19, 2009

One topic that is rarely discussed in both the academic and practitioner literature is the division between short-term and long-term market efficiency. The common assumption often drawn from research that implies markets are efficient is that this theory applies uniformly across the board to both short and long- term strategies. In theory, this is accepted to be true, however in practice this is far from the case because of the many structural factors that interfere with making markets efficient at all levels.

Here are the major agents that drive efficiency in both time frames:

Short-Term: intraday to 3 days: market makers, specialists, proprietary trading firms and trading desks, hedge funds, day traders

Long-term:1 month to 1 year): mutual funds, hedge funds, pension funds

Most of the liquidity in the markets in the short term are driven by computerized trading algorithms and this area is highly competitive. All agents can go both long and short with few restrictions, and have the ability to harness significant leverage to their capital to take advantage of discrepancies. In the short-term, the market is much more likely to be efficient and the commission/per trade return is higher making it even more difficult to make a profit. This is where the statistics that 95-97% of short-term traders lose comes in, because making a gross profit is not sufficient, you have to overcome the trade commissions as well.

At the long-term time frame we start bringing in the titantics of the world: large mutual funds, pension funds, and long-term hedge funds. At the fundamental level, these funds have the greatest leeway to make decisions whether they use qualitative (assessment of business strategy, management etc) or quantitative research (P/E, ROE, Debt to Capital, estimate revisions etc). This makes the fundamental game very competitive and also more difficult to make a profit. While alpha still exists using various simple fundamental data, it may be ephemeral, and I would argue having researched these variables significantly that the market is fairly micro efficient for highly liquid stocks. That is, selecting the best stocks especially using fundamentals is highly difficult. The only lasting alpha in the future will reside in effectively combining information from fundamental variables into a composite signal. 

However at the technical level, the possibilities are far more promising. Most of these agents cannot go short, and even if they do it is often for less money and on only the largest stocks or ETFs. They often have significant constraints on position sizing, and also on sector composition. Furthermore, they are significantly restricted from market timing and dynamic asset allocation in the conventional sense whether by regulations, or investment mandate. The percentage of funds that are permitted to perform these sorts of activities is very small, and they still don’t have much flexibility in making major bets. Thus in a macro sense, the market is very likely to be inefficient: ie the decision to allocate across industries, sectors, asset classes vs generating alpha within the same category. This is why relative strength strategies are so successful whether across asset classes or just market sectors, or even individual stocks. Jeff Pietsch of Market Rewind has published a good article on relative strength recently and just how durable it is over time:  http://marketrewind.blogspot.com/2009/10/two-simple-relative-strength-rotation.html Money has to flow somewhere, and ultimately it will flow to the most promising areas. Indeed the father of the efficient markets hypothesis himself: Eugene Fama has stated that momentum is the strongest and most robust anomaly in financial markets. Now you know why!

Well what about the intermediate term? Say 1 week to 1 month? I would argue that this is an area of inefficiency and a sweet spot for individual traders with a short-term horizon as well as medium sized hedge funds with  short-term horizons and complete flexibility. Most of my indicators are designed to target this area specifically because it is the easiest area to succeed in my opinion if you are a short term trader. The average holding period for DV2 and many others is between 5 and 20 days not coincidentally. To receive a very unique “Zone Report”  for the S&P500, Oil and Gold that tells you when and how to use various DV Indicators  like the supercharged DV (the more powerful version of the DV2 to be released) email us at dvindicators@gmail.com

7 Comments leave one →
  1. QuantPlus permalink
    October 22, 2009 11:45 am

    “more efficient in short term”. As someone whose made > 1,000,000 trades… it’s precisely the opposite. Virtually all pairs/basket trading and “reversion to mean” profits are made by using quant analysis to identify mispricing that corrects within hours or days. Why? Because is something does not “revert to mean” within days… you may be looking at a “paradigm shift” = “new mean” and will likely absorb huge losses. During the Financial Crisis… any strategy that relied on medium (weeks) to long term (month) “reversion to mean” was a suicide strategy. Those are the funds that blew up. Could not adjust to paradigm shifts.

    • david varadi permalink*
      October 22, 2009 11:52 am

      Ok lets clarify something……when i say short-term efficient, i am saying that it is difficult for the average person ie individual to make a buck in this time frame, obviously for those equipped with algorithms etc they stand a far better chance as indicated. The whole reason prop desks etc operate in this frame is precisely because they can do this stuff effectively with superior execution, lower transaction costs, and better access to order flow. Being someone that trades pairs myself, i operate also overlapping within this time frame–but i have better tools, execution and knowledge than most. Efficiency is a proxy for how easy it is to make a buck…..and clearly what you are suggesting is correct only for a handful of players.
      The largest players in the market like the mutual funds, pension funds, and individuals at retail full-service and discount brokers get killed in this time frame. Finally what you are referring to is pair specific also……and i agree that longer term pair strategies are riskier.

      best
      dv

  2. October 23, 2009 12:49 am

    The great equalizers across 5-20 day time frames are the commissions and slippage incurred from more frequent trading. Traders operating in this time frame need to research order routing strategies and strive to trade at liquid times, i.e., the close rather than the open, when possible. Also, commissions of .005 versus .008 can have a significant impact, even though we are talking about 3/10ths of a penny difference.
    I agree completely though, 5-20 is the sweet spot.

Trackbacks

  1. Short-term vs Long-Term Market Efficiency | Financial engineering resource center
  2. Tuesday links: making market calls Abnormal Returns
  3. Tuesday links: making market calls | Financial engineering resource center
  4. Trading News Center

Leave a comment