market vs. limit orders

"Tad Borek" wrote

[good til cancelled]

Do you measure an investor's performance by short-term, paper losses/gains? By this standard, many day traders are geniuses.

One cannot say the loss was guaranteed without exercising

20/20 hindsight. Or, in the alternative, you'd have to dress down the guy who was /watching/ the price and, necessarily arbitrarily, had chosen to buy at what he /guessed/ was the bottom of say $65 when in fact the stock fell to $64.6 before day's end. Oh the fool for buying at $65! (not)

The alleged "rational" order placer may just as easily guess the wrong price and either buy "too high" (by the 20/20 hindsight, short term, standards you propose) or not buy at all, because he set the price too low.

This is some of the stuff of some market timers. By contrast, I am interested in fundamentals and estimate a fair price based on them. When this price is reached, I buy. If the price goes somewhat lower, it's generally noise for the day, and I remember that generally I buy for the long run.

That paper promotes numerology and day trading, IMO.

"Elle" wrote

Oops. Tad gave an example that is far from typical of stocks; an instance where stockholders are about to be paid cash by a company and have no say in it.

I was speaking of the more general situations to which Beliavsky's paper refers. My remarks in the previous post apply to them. There are nearly always exceptions, but we are speaking of trends and general guidelines. In fact, the author of the paper B cites does not condemn per se the use of limit orders; he urges caution. The same caution should be provided for market orders, since at any point in time there is no telling what news is just around the corner to radically alter a company's valuation and/or stock price.

Forgive my ignorance, but I didn't follow this statement. I wasn't implying that all of the equity funds out there would fall under the scope of your stock screener. I was simply saying that, on average, historical data suggest the active investing has provided inferior returns to passive management. I was not questioning your method. It may work, but most methods do not.

Even if we assume your method unequivocably outperforms that still doesn't address the "everybody can make money this way" statement. The very essence of the efficient market theory is that if your method works, and everybody finds out, then everyone will begin using your method. As a result the knowledge of your method will become incorporated into stock pricing and eventually your method will no longer work. Its pure economics of demand pricing and free markets. Its the same reason that an infinite number of companies can't enter into a given industry and perpetually maintain the profit that the very first industry achieved (ceteris paribus, of course). Eventually competition, cost of inputs, etc, etc eliminate (or at least reduce) the advantage that the original companies had.

I also don't understand your argument regarding fees. Passive management can easily be done without large paying management fees. On the contrary some great index funds can be bought and held for only a few basis points. And as you said, that can be done without paying for screener software, analytical data, and brokerage fees. Surely you don't intend to argue that active management is more "fee friendly" than passive management?

Elle, you crack me up - sometimes I think you're intentionally egging us all on. The paper has about as much to do with numerology and day trading as it does with making a lemon chiffon pie.

If you don't see why you'd want to re-assess your limit order if a cash merger was announced at $10 below your limit price, I suggest you find out! It has nothing to do with day trading or market timing, it's a definitive change in the "fair price" based on some very obvious "fundamentals" (which change, as news happens...news like "this stock is never going to be worth more than $65 because that's the merger price").

-Tad

I read the book and have a low opinion of it. After making a big to-do about how P/E ratios alone do not predict stock market returns, Fisher eventually acknowledges that in conjunction with interest rates, they have shown some predictive power. One can search "fed model" at SSRN

to findresearch.

There have been many academic studies showing low P/E stocks outperform on average.

Come on, you've been reading my postings long enough to know that I know all that. Please reread Tad's original post, or even the part I quoted.

The limit order was triggered when a cash buy out was made at $65 for a stock that had been selling for over $69, thus the price fell and triggered his limit order. Yes, he was buying stock, but it was about to be converted into cash at less than his purchase price. Still not a road to riches.

-- Doug

That's exactly what I thougth I said (by "this strategy" I meant investing in small caps like the OP indicated). Small caps are historically undervalued at least in part because investors are aprehensive about investing in firms they are unfamiliar with (as opposed to blue chips).

I'm not plugging a strategy. I don't actively invest or use a "method". I firmly believe in asset allocation and buying/holding. In the post above I was saying that I believe active strategies may work, but not on a large scale and not once they become widely used (as per the efficient market theory).

Yep. It can be both entertaining and frustrating to see the market grossly over-react to to new information.

Thanks for the reference. I will be sure and take a look at that.

"Tad Borek" wrote

Sorry you don't get it. The paper obviously completely ignores the conventional wisdom of investing in stocks only for the long term. It's very much about short-term trading. More grist for the academic mill, with its only practical salient point being (for all buyers, limit or market): Keep an eye on the companies you plan to buy/sell.

I explained my misreading of your example earlier. The misreading is easy because your example is so extreme. Preliminary announcements and chatter about possible buyouts are the rule. The sudden price adjustment you propose is unlikely, unless its related to a panic of sorts (which are common). Or only someone completely oblivious would fall into the trap you propose.

Fact is anyone buying or selling stocks should be informed, and the same caution should be applied whether they use a limit order or a market order. You may not like GTC orders, but they are absolutely no big deal for long-term investors who exercise reasonable caution. Arguably they represent the advantage individual investors may have over mutual funds, since the fund managers are not necessarily shopping for a good price on a stock.

I think by "work" B meant beat the indices. In which case, many stock-picking strategies do not "work" - witness actively managed mutual fund performance vs. indices.

-Will

Isn't asset allocation an active investment strategy?

-Will

I guess we could argue definitions and technicalities but I think the short answer is no. At least not in the sense to which I was referring. I buy once and rebalance annually, that's it. According to investopedia, this is a "buy and hold strategy" which is often associated with passive investment. Investopedia acknowledges there are asset allocation methods (like Tactical AA and Dynamic AA) that more resemble active investing however I do not recall anyone on MIFP ever referring to those methods when discussing "asset allocation", nor was I just now.

Will, your question implies that if you ever buy and or sell a stock other than the original stock purchased you are engaging in active management. Is this really your perspective or have I over-read your statement?

Yes and no. It seems to me that active investing depends on making more than trivial decisions on when to buy and sell. Passive investing implies using an index fund (I realize that these are not conventioanl definitions of these terms). So the arbitrary decision to buy two index funds and allocate say 80% of investment funds to an S&P 500 index (for example) and 20% to a total bond index (for example) requires an active decision. So what? This is probably a trivial decision, so still should be considered as non-active investing.

However, as I recall from your previous posts, you use a risk-based approach to asset allocation. Meaning that you are using a non-arbitrary method of determining your allocation, attempting to get close to the efficient frontier by using asset volatilities and expected returns. Presumably this allocation will shift over time as new data comes in and will be affected by what I would consider to be non-trivial decisions about your model. Even if implemented with indices, it seems that this strategy is at least as active as say a strategy that overweights the lowest P/E stocks in an index and then reallocates once a year.

I apologize if I have misrepresented how you allocate, but the above makes a good example for discussion nonetheless. Perhaps allocation as described above should be considered "mechanical" as opposed to "passive." If true, I would then posit that any mechanical system that does not involve frequent trading is passive. Active strategies would then be those strategies that are not mechanical (or, thus, passive). The strategy presented by the OP would then be essentially passive (for purposes of discussion).

Maybe.

-Will

No need to apologize. I think you are correct in your assessment of how we/I invest. We use about a dozen ETFs that encompass something like 99% of the market thus closely representing an efficient frontier. Based on the investors risk tolerance we then move them along the efficient frontier to determine the percent composition that each ETF will have in the portfolio. We then rebalance back to those %s annually. You are absolutely right that if one's risk tolerance changed we would change the composition %s of the portfolio. Realistically, we only make minor adjustments every 3-5 years. Of course, if a major (usually unexpected) life event occurs, an investor may find that his/her risk tolerance has radically changed over-night. In that event we would likely change someones asset allocation even if we had rebalanced the day before. The ETFs do not change, but there % composition certainly could/does.

Maybe that is active trading by some definitions. I have never really considered it thus because the investment changes are not made based on the fundamentals of the investment or current market trends, but rather the needs of the client. I think your use of "mechanical" is pretty good way of looking at it. Under that rationale I also think its possible that the OPs method could be mechanical/passive, but its pushing the limits (there is alot more lilkelyhood of trading in and out of positions to exploit pricing phenomenons). I don't personally have a problem with the OPs methods. The thesis of my early posts is that based on EMT and demand-pricing economics, ceteris paribus, ANY method that attempts to take advantage of investments based on their over or under valued status MAY work, but if they become widely know, understood, and/or engaged-in their success margin will decline, possibly to negative numbers. Think about how much more success an investor would have if he were the only person in the world that knew about the low p/e phenomenon. Less people competing for those stocks means lower purchase prices. That's why I contend that every investor out there, even the uninformed, could not realistically make money this way.

The page links further to a definition of Active management, which offers "The opposite of active management is called passive management, better known as indexing."

Now this would strike me as suggesting that any definable, rules based system which at any point can define the stocks to be purchased, can be forged into an index fund, and would be considered passive. And asset allocation, which is rules based, to adjust back to a given percent mix either based on the time or deviation from the initial target, would be no different.

Am I off base here? JOE

How would you classify investing using a defined strategy with yearly re-balancing; such as the various strategies discussied by O'Shaughnessey in What Works on Wall Street?

--ron

Both of you have equated "passive" with "mechanical". I don't agree. A strategy that bought at the close each day stocks that had fallen from the previous close would be mechanical, but since it would have turnover exceeding 100, it would not be passive. The strategy I mentioned of buying low P/E stocks and holding them as long as they remain low P/E is somewhat active, because stocks would be sold when their P/E's rose too high.

Yes, but that does not mean that the next reader of MIFP that tries to do so will fail. My point is that the "barrier to entry" of a systematic strategy -- one that can be computerized -- is lower for many individuals than the traditional qualitative approach requiring in-depth knowledge of the fundamentals of hundreds of companies.

Your approach of periodic rebalancing of asset classes is reasonable, but it could *not* be followed by all investors. Suppose that the dollar-weighted optimal allocation of all investors is 60/40 stocks/ bonds, but that ratio of market caps of stocks to bonds is not 60/40. Clearly not everyone can have the optimal allocation. Looking at the problem dynamically, suppose that in a bear market the market cap of stocks falls by 40% while the market cap of bonds rises. Your rebalancing approach would tell people to sell bonds and buy stocks, but if everyone rebalanced, who would be there to buy bonds and sell stocks?

Neither your method nor the low P/E strategy can be profitably employed by every investor, but that does not mean that either approach is wrong for the *marginal* investor.

I don't really know what I think yet (I thought I did, but now I am re- evaluating my position). Like I said, I have always thought of using index funds to create an asset allocation as being passive. I think perhaps Tad did a better job of explaining my stance than I have so far. Its not necesarily the investments or the frequency of investment (within reason) but the reason for investing that differentiate active from passive.

I think this is what I said when I wrote, "I would then posit that any mechanical system that does not involve frequent trading is passive." No? Further, a system that depended on non-mechanical analysis of securities would be active as long as the turnover is significant. I add the turnover caveat since at least one "index", the DJIA, is essentially actively managed - these stocks are selected. But the index is changed infrequently, so that a fund based on the DJIA would be a passive fund, not an active fund IMO.

-Will

I agree, that's why I had the caveat "any mechanical system that does not involve frequent trading" in the part you quoted above. My point was that if your strategy was implemented with trades once per year (although you stated in your last post that trades could occur more frequently) it would be no more active than a risk-based asset allocation stategy that rebalanced once per year. Thus if the asset allocation strategy is passive, then I would think that the low P/E strategy could be considered passive. I don't think either are particularly passive, however. But the distinction is not that important to my point, just the relative equivalence of the activeness of the two strategies.

-Will