The nature of investment decisions and portfolio management

 

There are several ways of conceptualising investment decisions –

 

-         Buy a perpetual income stream cheaply (the ‘value’ approach, or perhaps the ‘actuarial’ approach)

-         Buy in anticipation of a change in market perception of the stock (the Keynesian beauty contest metaphor)

-         (a novel idea, and the subject of this note) Equities as state-dependent securities => buy in anticipation of a favourable change in state.

 

 “Buy a perpetual income stream cheaply” is the most intellectually appealing in principle, but always impossibly difficult in practice.

 

“Buy in anticipation of a change in market perception” raises the problem of why you should, on average, have superior insight into future market perception.

 

“Equities as state-contingent securities” refers to the idea that each equity is a piece of paper which has different payoffs in different states of the world. “Buy in an anticipation of a change in state” is a helpful concept because it highlights that

 

-         you don’t need to know everything about a stock, you only need to see some future change in state (particularly one affecting the financial dynamics of the company’s share price – Rushbrook)

 

-         you often aren’t trying to “value the company” absolutely – just focus on possible changes in state

 

-         you need to focus on knowable changes in state (eg the anticipated completion of a takeover, a change in legislation, etc)

 

-         this is a way of allocating research time – ie focus on looking for things which are knowable, and forget things which aren’t (investment as applied epistemology!!)

 

-         mean reversion (eg of dividend yield) could be regarded as a ‘change of state’ – in this sense the approach encompasses value investing.

 

-         Stockbrokers’ reports, and most other commentary, and most investment managers, do not have this focus on knowable changes in state.

 

There is still a problem: “Why should you, on average, have superior insight (as compared with other investors) into future changes in state?” The answer is that you probably don’t – but what you can have is selectivity – ie you bet only when you do have superior insight, when you anticipate a change of state with high degree of confidence.  (“The vast majority of stocks are not compelling either way, so ignore them” – Buffett.   “Patience, coupled with decisive action” – Munger.)

 

Avoiding mistakes is also very crucial to investment management, because of the ‘arithmetic of losing’ (ie down 25% requires up 33% to recover; down 40% requires up 67%, etc). 

 

As a corollary: type 1 errors (reject a good investment – an error of omission) are generally preferable to type 2 errors (accept a bad investment – an error of commission).  Note that reducing errors of commission will probably increase errors of omission – but because of the arithmetic of losing, that’s probably OK.

 

If you can avoid mistakes, outperformance may happen without any particular positive decisions being responsible.

 

The need for knowable things may lead you into shorter term investments, and hence possibly higher turnover and tax (“when the state change has happened, sell”).  I’m probably prepared to accept this (as per Buffett 1965), although I’m also very aware of the tax attractions of long-term holdings.

 

After the “change of state” occurs, there may not be enough reason to sell (especially if it was just mean reversion).  It may be acceptable to continue holding without a foreseeable change of state,” if the company has high fundamental quality (eg ROCE, operating margins, management discipline, etc).

 

Often valuable to decide to do nothing – accept time being spent on this.

 

Fully rational optimisation is usually an impossible aspiration in investment, because of the complexity of the problem.  Heuristics can often be more useful.