[Guest Post] Capital Allocation in a Portfolio

This post is from Prabhakar Kudva (http://investment-in-sight.blogspot.com). 

Capital allocation is probably one of the most important aspects of investing. Every time I need to decide how much capital to allocate to a particular company i use the following steps:

Step 0: Identify the companies whose business dynamics you understand reasonably well - either because its inherently a simple business and/or because you've spent time and energy to understand what factors affect a particular business' performance. This is basically your universe of companies.

Step 1: Predict the approximate EPS (a range,may be) one year down the line.If you are unable to predict the EPS with a reasonable degree of accuracy then it means two things:
a) This is a complex business where profits depend on a number of totally unpredictable factors.Its better to remove such companies from your universe/sample space.
b) You don't understand the business well enough.Read more about the business.Do some scuttlebutt.This is a learner's game.If you spend time understanding simple businesses, their sources of profit and external factors that affect business, eventually you'll be able to estimate the one year forward EPS with a reasonable degree of success.

Step 2: See if there is a possibility of a PE re-rating. Or will the PE remain the same. Or may be the PE will be de-rated? Estimate what the PE might be based on your projections in Step 1.Remember to be conservative.In most cases assume PE will remain the same or there'll be a slight re-rating(if expected profit growth in step 1 is out of the ordinary).If you think there's going to be a de-rating you know what to do.

Step 3: Now you have an approximate PE and an approximate EPS range. Arrive at your target price.

Step 4: Now based on the CMP and target price - check what the expected return is.

Step 5: Repeat steps 1 to 5 for all companies in your universe (step 0). Compare the expected returns arrived in step 4. For example if you have three companies A,B,C and the return expectations are 40%,25% and 10% respectively you should invest:
a. 40/(40+25+10) = 54% in company A
b. 25/(40+25+10) = 33% in company B
c. 10/(40+25+10) = 13% in company C

Assumptions
a. Ofcourse this is not exact science.Every year that you repeat this exercise,learn from your mistakes, learn new things about your business you'll get better at predicting the EPS of the businesses you understand and hence better at capital allocation.
b. I use just one year because I feel predicting more than one year ahead for ANY business is futile. The business environment and the capital market that we operate in is way too dynamic to talk about the 'really long term'.So we take one year at a time.