Positivity in Negativity
Nothing is ever quite as it seems in investment, it always strikes me. While I'll often say that I think things can be viewed from perspectives that are contrary to prevailing opinion, I often wonder just how far I can take that premise - looking for the good stories in bad news - without getting into muddy water and being a contrarian for the sake of it. After all, I don't disagree with things for the sake of disagreement; I'm trying to beat the market. You can't beat the market by agreeing with the market. Bear with me, then, as I'll restate one of my usual 'good-spin-on-bad-story' examples and try and push the principles to the limit.
The story I usually bring out is the one that's most familiar to me; mortgage approvals. I've included a graph below which shows the number of mortgage approvals from 1994 to now, with statistics from the Bank of England.
The question you have to ask is whether the current low level of mortgage approvals is a bullish or bearish sign for the housing market. On a side note, it should also make last month's rather lamented 'dip' in mortgage approvals appear rather insignificant in the grand scheme of things, as well as well within the ranges of what appears to be normal noise. The same applies for when it jumps upward for a month and people latch onto it for hope! Decided on the bullish vs. bearish indicator, then?
I've probably said before I think low levels in an indicator, perversely, are often a good sign. After all, if house prices are stable with a low level of demand being exhibited in the market - relatively stable at low levels of confidence - I would say the balance of risks points to a recovery in the housing market. I think the chance of confidence improving is better than the chance of confidence decreasing from its current low level.
However, I do think I must be logically consistent. Let's park ourselves in 2007 on the graph. Mortgage approvals are at a historically high level. My viewpoint - that the current low levels of mortgage approvals indicate a lot of upside and relatively little downside - is only coherent if I argue that in 2007 I would've been worried about the market being overheated. Would I have the confidence to make that call in the face of consistently rising prices? At what point do I decide things are getting dicey; 2005? 2006? 2007? Difficult questions, but they need considering. The argument is essentially that's in the change in lending, not the level of lending, which should be considered.
It's exactly the opposite of the optimism that many experienced (including some of the media) in the mid part of the 2000s; mortgage approvals are still high, confidence is high, lending is flowing - prices will continue to rise. They are being consistent by now arguing the opposite; mortgage approvals are low, confidence is low, lending is jammed - prices will drop. The problem, it seems to me, is that they are perennially behind the times. When you're noticing that things are incredibly rosy, I reckon it's time to proceed with caution, not time to predict further gains. Time will tell if I can use that conviction to any good use or if I'll get caught up. I expect the latter; it is, like most of the cognitive biases, rather easy to point out but crushingly difficult to avoid!
Anyway, that rather meaty thought aside, I wanted to tie up two concepts; the above, and a post on iii blog about looking for a better indicator of firms' profitabilities. They may seem disparate, but my interest comes from a different angle. I really like the premise - cutting through the dust and smoke and getting to what really matters underneath all the obfuscation of accounting principles and capital structures, but reading it sparked me off a little about one of the building blocks Richard was using for determining if a firm was consistently profitable - a long term calculation of return on equity. The benefit is fairly clear - if you can put in a conservative and reasonable RoE figure, determined from a history of performance, you can calculate a reasonable price to pay for the company with more conviction than by using just one or two years' earnings, which are subject to some noise and may just be particularly lucky.
My question is a simple and open one, and I'll leave it without too much discussion as I am genuinely unsure. Several academic studies have suggested that RoE/RoA/RoI (and other similar figures) are, at least to some small extent, mean reverting. Mainstream economic theory would suggest that this might be the case; after all, it is supposedly excess profits and high return on capital which encourages competitors to enter the market, and drives returns down to normal levels. Returns on capital are normalised around the economy and money flows to where it earns most.
In that way, a successful investment strategy might actually be exactly the opposite of what you may normally think. Similar to low mortgage approvals (downside is limited), is there any credit in a strategy which targets cheap and financially sound firms with a low RoE/RoA? Perhaps there is low hanging fruit they can grab at to bring returns up to their competitors' levels. Perhaps those firms with high RoEs are benefiting from factors which, in a free market, will prove to be transient. Perhaps it's just a way of justifying buying bad firms because of the allure of low prices! If mean reversion holds, though, it may have some credit.