## \$TSCO.L – my valuation: 145p using CAPM, and a completely blunt stick

With TSCO (Tesco) in the news, it’s worth asking the question: what’s it worth?

In March 2014, I valued Tesco at 360p (http://is.gd/wnWgzZ)
In August 2014, I came up with a valuation of 200p (http://is.gd/EyL7c9)
I’m now valuing it at 145p – a significant reduction from my previous valuations. Erstwhile, I had been using the Gordon Dividend Growth Model. It’s a nice simple model, but I think it is more difficult to apply to Tesco’s current situation. I will therefore adopt a more complicated model for this exercise.

In this valuation, I am taking a chainsaw to anything that will complicate my life. It is a very quick, and very dirty, valuation. What I should do is an elaborate DCF calculation, but I am far far too lazy for that. I’ll be using interest rate and equity risk premium data from November 2014, and assume a beta of 1 for TSCO. Anyone wanting to do a proper valuation will be up in arms about this. So when I’m saying I’m doing a quick and dirty valuation, I’m really not joking.

However, I can really simplify my work.

Suppose I have an operating profit for a company, a dividend cover, and I know its ROCE. Let me suppose that for a period of 5 years, the company reinvests at a rate implied by its reinvestment rate, and then adopts the average of the market thereafter. Given that I am making so many assumptions, it should be possible to somehow scale the operating profits of a company by looking up some appropriate values in the table. As luck would have it, I have already composed such a table. It is available here: http://pdf.markcarter.me.uk/aswath.pdf

The PDF contains quite a few formulae, and allows you to tackle the general case, but what we’re interested for this exercise is table 2, for the “opscale” (operating profit scale).

According to Stockopedia, TSCO has had an average ROCE over the last few years of 11%. Let’s call it 10%. It’s difficult to say what the dividend cover over the next 5 years will be, but let me take the average over 5 years which include analyst forecasts of what the cover will be. The mean is about 3.5 (average of 2.66, 1.29, 1.61, 6.92, 4.9). The last two numbers are analyst estimates, so you can see that they are predicting very few dividends for TSCO.

If I look at table 2 in the PDF, that implies that I should scale operating profits by a value of around 11 to obtain the value of the firm.

What operating profits should I use? Well, according to my calculations, the operating profits for TSCO over the last 5 years had a mean of 1404m and a median of 2354m, which includes the recent horrendous operating loss. Let’s split the difference, and say that the operating profit will be 1879.

So the value of TSCO as a firm is 1879 * 11 = £20669m. However, TSCO has substantial debt, to the tune of about 9000m. I have to deduct this from its firm value to obtain the value of equity. So, TSCO equity is worth about £11669m. It has 8123m shares in issue, giving a fair value for the equity of around 145p.

This would suggest that TSCO, at 225p, is still overvalued. Arguably, if you’re something of a pessimist, operating profits and ROCE will be lower, suggesting that even the 145p needs to be knocked down somewhat. Maybe you think the dividend cover will be greater for the next five years, in order to give TSCO a bit more of a cushion. In that case, the valuation may creep up a little. You will notice in the table, though, that dividend covers don’t start having a more profound effect on valuations until the ROCE is higher.

You can play around with these assumptions to be more to your liking. To be more accurate, you would really need to start looking at proper hurdle rates and time periods. It is likely, for example, that due to the status of TSCO’s bonds, the hurdle rates should be higher than the one that is implied in my model. You will really have to roll up your sleeves and do a proper DCF to see what effects that would have on the valuation, though.