Trackers traditionally outperform in bull markets so investors looking to position themselves for recovery should buy now. But getting a fund to track the index accurately is not as simple as it sounds
After several years of equity market declines, it is hard to see why anyone would want a fully invested equity product. Yet this is precisely the time when index funds might be a shrewd choice. Statistical work done at Standard Life Investments shows that in a bull market, index funds outperform the median active manager.
Conversely, in a bear market, the median manager outperforms the index fund. This is what has happened in the current bear market. If investors are to position ourselves for a market rally, then being fully invested with an index fund is likely to be a winning strategy. Index funds are a bull market tool.
As active fund managers push active risk levels ever higher to try to capture more alpha, the inevitable consequence is that active funds become more concentrated. The typical long-only position is several percent of the fund, far higher than the index-neutral position.
Should a stock story go wrong, the fund will be torpedoed, and it will dramatically underperform the index. Similarly, there are likely to be several large underweights in an active fund, stocks for which the fund managers have deeply held beliefs the market will tumble to a negative story. The risks here are that the 'dog stock' gets a takeover approach, and the stock price rises.
There is a third risk also. Concentrated funds are comprehensively underweight the smaller end of the stock spectrum. The number of holdings tends to be very small in comparison to broad indices. If a collective correlated phenomenon occurs (like for example the internet bubble) then lots of tiny stocks begin to act as one, and rise in the rankings. Then the many tiny stock underweights come together to make a large underweight theme. Again the fund is hurt, by this emerging theme. In all three cases, an index fund, by definition, is neither overexposed nor under exposed to the risks.
One common misconception is that index funds are buy and hold, and thus pretty simple and cheap. This is not the case, as I will outline here.
In reality there are relatively few design choices that need to be made. The key decision is how good a match to the index you want. If all you want is broad index agreement, then cash plus futures, or a representative basket of stocks, will do. This gets close to a tracking error of 1% with respect to the underlying index. While this may be good enough for temporarily soaking up spare cash in a larger, nearly fully invested, portfolio, it is not a credible index fund approach in itself.
The mid-range solution is to optimise a basket of stocks to a benchmark. This equates investment between the index and the fund along several obvious key dimensions, such as industry classification, and some less obvious ones like style. Typically, a reasonable index fund can be built with 500 stocks optimised to a benchmark of 720 stocks. A typical tracking error of around 30 or 40 of basis points results. Moreover, a risk lurks herein.
If you look closely at optimised portfolios, you find that the fund has good representation of the largest 350 stocks, but becomes patchy as 150 stocks have to proxy for the smallest 370 stocks of the index. This means that an optimised fund behaves like an active small-cap portfolio. So it follows that the emerging theme risk is still present. Indeed optimised trackers performed badly during the internet bubble. Optimised trackers are not really credible any more as marketable index products.
The full-size solution is to fully replicate the index. This can get to tracking errors below 10 basis points, which at first sounds easier than to optimise. Less computation is needed and it is more accurate. However, this accuracy comes at a price. We now have to follow each and every index action slavishly to ensure that the fund follows the index. It is an interesting thought that below the placid surface of the index lurks a seething mutating mass of corporates doing index altering events to each other. Thus we have to work at it.
It becomes an open-ended problem to ensure that every corporate action is dealt with on information gathered before hand and in a way consistent with the index. This has become more complicated in recent years, as options for deliverables on accepting an offer have multiplied. The response from the active market to the options will determine what the index committee eventually decides to reflect. This is where fund managers have to be careful. It is important to have access to the index changes before they have been implemented, which means having good contacts with index providers, and with many brokers who have units which follow index changes.
There are many other techniques to ensure that index funds are managed efficiently. For example it is good practice to keep of the order of 1% of the fund in futures plus cash to allow index events such as a stock inclusion to be funded with one sale line only, rather than 720 thin shavings of a whole fund. Dividend flow has to be reinvested on 'ex-D' date rather than when the cash finally arrives.
Are enhanced index funds the way forward? This is an area which is beginning to squeeze traditional plus 1% mandates for institutional fund managers. There are many tools used for enhancing, but I will only focus on one here.
A common way to enhance an equity index fund is to avoid the trading date for index transitions. If a large stock is to come into the index, then the stock ought to be bought just at the close of the market the previous day. Then the fund and the index will behave as closely as possible, and the timing risk of the stock purchase for the fund will be minimal. Clearly if managers follows this rule slavishly, then they are a ready target for less rule-bound investors, who may choose to move the price into the close such that the index fund pays a silly price.
This classic effect hit a peak with the inclusion of the software firm Dimension Data into the FTSE 100. The price spurted from roughly £6.65 to £10 during the final minutes of trading before the index inclusion, and then fell back to the £6.60 level the following day.
Thus funds which traded on the close may have matched their indices, but anyone who traded at a different time made money relative to the index.
This kind of effect suggests that timing index purchases is a money making strategy. There is clearly a spectrum of timing from a few minutes to several days, and the longer the timing mismatch the higher the risk on the trade. Indeed, the index fund may break its mandate if it takes too much timing risk on board. This is particularly true now. Index transition events are so well flagged these days that the events are overtraded. Too many investors think they will be able to load up on a hot stock early and make use of the guaranteed exit to an index fund at the transition point. Recent major index trades have been better traded at the proper time.
Clearly there are other strategies for enhancing an index fund, but they all involve taking risk, which can be ill-advised. Enhanced index funds can lose money. We tend to trade close to the transition point (we are tight trackers) because that is what the mandate requires us to do. The only exception is when we have good evidence (from information flowing into the active fund managers we communicate with) that a market squeeze is imminent. We did not buy Dimension Data at the close.
In conclusion, we have demonstrated that index funds are actually quite complicated when run professionally. In particular index changes and corporate actions need careful treatment. Funds which do take the required care, will reward investors with a clean, fully invested vehicle that will capture all of the upside in the market, when the turn comes.
Active funds generally perform relatively poorly in a rising market.
Different types of trackers track the index with different levels of accuracy.
Trying to time index moves can increase risk and may break the mandate of a fund.
'Life catches up with us in the end'
‘Personalised Predictive Analysis’ tool
Summer series continues
Both start in August