No. of Recommendations: 21
DTB wrote:
If you decide you want to reallocate capital from a company with a high valuation, relative to your estimate of its value in 10 years, then one might wonder how finely tuned such a readjustment should be. For instance, if a stock is trading at 5 times your estimate of its intrinsic value in 10 years, and drops to 4.9, it is a slightly less good deal, and it would make sense to me to very slightly reduce exposure to it, reallocating it to something that has maybe gone from 4.9 to 5.0, for instance. And then again, if it drops to 4.8, an so on. The alternative would be to wait for a much bigger drop, like from 5 to 4, before reallocating. I'm guessing the former is a lot more trouble for about the same result, but I think you might theoretically gain a small amount from doing the rebalancing more frequently. Maybe not enough to be worthwhile, but positive nevertheless. But I would be interested to hear your opinion on this question.
This framing of the question shows your excellent familiarity with the Manlobbi Method. You are aware of this, but to remind others the intrinsic value in 10 years (as relates to the IV10 term) is the lower bound for the range of expectations of the intrinsic value (the value of the firm itself plus everything paid out) rather than the central estimate. Consequently for unpredictable firms, there is a wide mist of outcomes and you end up appraising a pretty poor IV10. So the high scoring IV10/price ratios are nearly always with unusually predictable firms (predictable often only for yourself owing to peculiar insights, for which others will objectively classify as unpredictable, and thus should rationally be skipped as non-Steadfast). For some situations it really is possible to conclude a fairly narrow range of future 10-year intrinsic values. To see this intuitively, value your house in 10 years (exit value plus all the rent paid out, assuming no mortgage). You will find that your range of possible values is amazingly narrow. Some firms are no less predictable once you understand them economically, and their long-term cultural/operational principles. Don't worry about inflation - simply assume some fixed value because we are only ever comparing IV10/price ratios for different firms and the objective value has exceedingly less efficacy - it is all about the relative IV10/price ratios that we care about, and what governs our actions. Stocks, bonds, index funds, real assets, even small private business ventures, all have an IV10 value provided the Steadfastness filter is first passed.
Secondly, you don't want to want keep allocating capital from an expensive firm to a cheaper firm indefinitely, or you could fall into trouble if that firm really has problems and you are moving capital into a slowly disappearing void. That should never happen if you were truly cautious with Steadfastness, but it must not be ruled out. Conventionally one limits the maximum size of each portfolio position, but that completely fails because the position can keep shrinking, and sucking up your capital as you buy it at lower and lower prices. Thankfully this situation is easily averted completely by the following rule, which amazingly I never see articulated anywhere but here it is: Limit the historical amount of capital sent to any one position to a fixed percentage of your equity, rather than limiting the destination allocation. As a concrete example we set the maximum historical transference amount as 20%. Next imagine that you purchase a 10% position in XYZ. It then halves to 5% with the share price falling by 50% and you enthusiastically allocate 5% to it, so now you have allocated about 15%. Then it halves again to 7.5%. You really want to buy a lot of it now, but you have already allocated 15% and you can only allocate another 5% and then that's it - after that you are no longer, no matter how many years pass, able to allocate more capital to it.
With that, we can get to your question. In Manlobbi's Descent I wrote that the difference in IV10/price between the expensive and cheap situation should be at least 1.5, though in practice you want to aim for a delta more like 3.0. You might sell one firm with an IV10/price of 2.1 and want to sell some of that and buy another one with an IV10/price of 3.3 (a delta of 1.2) but you should be patient and wait for the IV10/price difference to be larger. Part of the reason is the previous paragraph - we want to wait for the larger deltas and strike then, rather than striking too early and losing our future opportunity to exploit larger price difference. There is some mathematical rigour behind this also. The benefits of equal-weight balancing between semi-correlated stocks gives a much higher CAGR benefit the more they are uncorrelated (the larger the IV10/price delta between two situations for which a trade is executed).
To answer your question extremely shortly, I would only buy and sell between two situations when the IV10/price difference is around 2.5 (and certainly more than 1.5), and I would strictly apply the 'historically allocated capital limit' rule above also (I regularly go all the way up to 50% provided these concentrated investments are for firms in the Steadfast universe). With enough time passing, and following these rules, you will outperform the market even more than the positions already outperform owing to the high IV10/price ratios themselves.
- Manlobbi
