One growth stock on the FTSE AIM All-Share Index I’d buy, and one I’d sell

Nichols (LSE: NICL) and YouGov (LSE: GOV) are among the larger stocks on London’s junior AIM market. They’re well-established companies and I view the attributes of both their businesses as highly attractive for investors.

Nichols owns a strong stable of soft drinks brands, headed by its flagship Vimto range. A recent escalation of the conflict in Yemen has halted shipments to that territory for the time being but the company’s geographical diversification makes this a relatively minor — and possibly temporary — issue.

Pollster and data products and services provider YouGov has similarly attractive credentials and, like Nichols, has been growing revenue and profit well ahead of its market.

However, while I’d buy one of these stocks on the Warren Buffett principle that it’s “a wonderful company at a fair price,” I’d sell the other, because I believe it’s significantly overvalued.

Getting the measure of earnings

The table below shows earnings per share (EPS) for Nichols’ last five financial years (year-end December), EPS being the ‘E’ component in the widely used price-to-earnings (P/E) valuation metric. I’ve included the company-adjusted EPS, adjusted EPS calculated by independent analysts at Morningstar and statutory EPS.

As you can see, the company and Morningstar adjusted EPS are very similar. This is because they calculate the number on more or less the same basis, which boils down to excluding exceptional items from statutory EPS.

Thus, one-off management restructuring costs and a provision for potential costs of a litigation claim were excluded in 2013, the costs of an award against the company for damages were excluded in 2014 and a gain on the acquisition of an associate company was excluded in 2016.

Based on a share price of 1,575p and adjusted 2016 EPS, Nichols’ P/E is 23.8.

An earnings magic trick

The table below shows EPS for YouGov’s last five financial years (year-end August).

As you can see, comparing the Morningstar adjusted EPS with statutory EPS shows that YouGov has one-off costs every year. Even more striking are YouGov’s own adjusted EPS numbers. Why are they so much higher than Morningstar’s? Well, in addition to exceptional items, the company excludes a whole host of other costs, the most pernicious of which is amortisation (6.2p a share in its latest financial year).

YouGov doesn’t expense its significant costs of developing and acquiring software, data and suchlike but puts them on the balance sheet as intangible assets and amortises the costs over a period of years. By neither expensing these very real costs in the year they were incurred nor including them in EPS in subsequent years, YouGov simply makes them disappear into thin air.

Based on the company’s adjusted 2017 EPS and its recent share price of 315p, its P/E is 28.9. This is well above Nichols’ 23.8 anyway, but using like-for-like adjusted EPS (that’s to say, as calculated by Nichols and Morningstar), YouGov’s true comparative P/E is 67. Remarkably, in their forecasts, City brokers go along with YouGov’s methodology. I view this methodology as little more than a charade and rate the stock a ‘sell’, based on its P/E of 67.