Restaurant stocks can definitely drive you crazy. On the one hand, there are reasons to believe these can make a tasty long-term addition to one’s portfolio. But we can just as easily make a case for the idea that too many of these are indigestible. Perhaps we can solve the dilemma by seasoning with a dash of fundamental data. (Pardon the puns; I covered the group for a while in the last century while working at Value Line and can’t resist.)

^{Is it better than what you can cook on your own? Photo from: JIM WATSON/AFP/Getty Images}

The Case For Restaurant Stocks

People are busy. The idea of sitting down for three square home-cooked means per day is fine for TV Land and other nostalgia-oriented programming services, or for those wealthy enough to have personal chefs. Most of us, however, eat on the go and often eat badly because shopping and cooking take time we can't spare.

So we need alternatives. Eating out is one of them, especially for lunch, when many are least likely to be at or near home and we’re seeing more and more of that for breakfast too. We often have more options for dinner. But who wants to choose home cooking? We aspire to imitate TV chefs, but in reality, we often wind up with home-cooked slop or a plastic container with of sodium garnished with microwaved meat and veggies we bought frozen at the supermarket or convenience store.

So there is definitely a place for restaurants of all kinds depending on one’s tastes and budget.

The Case Against Restaurant Stocks

We live today in a world of choice. We choose how we shop (on line or in store). We choose how we read (paper or screen). We choose how we communicate (too many to enumerate). We choose how we consume entertainment (too many more). We choose how to call for cabs. We choose and choose and choose, and we choose how to eat what we don't cook.

We can go out to restaurants, just like we did when mom or dad were at the steering wheel. We can pick something up and bring its home, in many cases without leaving our cars. We can call for home delivery. And, lately, we’re even able to make finer choices for home delivery; we can order chef prepared meals from one of those new on-line services (Cook Unity is big in my neighborhood). We can have pre-chopped and sorted ingredients delivered to us with prep instructions so simple even we have a 50-50 chance of producing something edible. In some areas, we can have Uber deliver our meals and even lately, amazon.com.

So as much of a place as there is for the traditional restaurants, size of the pie is shrinking as other alternatives move in. In other words, the C-word: Competition.

There’s another C-word to fear: Costs. Minimum wages are rising and that’s a big item for restaurant operators. Meanwhile, they have to figure out how to fairly compensate critical kitchen staff that can’t share in tips. And of course there are food costs and occupancy costs.

Meanwhile the margin for error is shrinking. To heck with the celebrity food critic. It’s the Yelpers who can make or break a restaurant.

Tastes and preferences are changing too. Big national chains are in (lower cost, better consistency, more convenience). Big national chains are out (less able to respond to local preferences which can get pretty exotic).

And The Winner Is . . .

Bullish? Bearish? How about: Both of the Above.

I think the argument in favor of a stake in the restaurant industry is too sociologically compelling to justify out-and-out avoidance.At the same time, I see the challenges as too pressing to simply pick a stock, any stock, or to naively run after the best earnings surprise to have been reported in the last quarter.

We need to figure out how we can identify firms most likely to be able to roll with the times and adapt to what the world throws at it.

It’s tempting to do the so-called Peter Lynch thing (get out there, look around and invest based on what you see in day to day life). But I’ve seen that such micro samples can deceive, especially in a group like this, which depends so much on experiential factors. So as cold hearted as this may seem, I see the challenge as one that may be best approached through cold hard fundamental data.

Searching for Tasty Restaurant Stocks

I found I can use the same screening techniques I last wrote about in connection with finding traditional retailers.

I went back to Portfolio123 and this time focused on restaurants for the screen I built based on characteristics I assume we'd see in a publicly-traded chain that wasn't dying.

Again, as with retail, I’m accepting the notion that change will occur and am looking to separate potential winners from potential losers by aiming at the kinds of metrics most widely used for this sort of thing.

As before, I hone in on return on equity (ROE), perhaps the single most-telling measure of fundamental merit we have. (It’s hard to argue against the idea that anyone who can turn $100 of capital into $20 profit is doing better than another who turned that same $100 into $6.)

I focus on the recent past, the period of time in which the industry’s woes has been especially impactful. My screen requires that ROE be positive for the trailing 12 month period, and the 12 month period before that. Next, I check for trailing 12 month ROE that is above ROE for the prior 12 months. These tests eliminate a lot companies. And if we really think about it, the ones left behind are more likely to be the ones we want to be looking at. Negative or declining ROE is a clear symptom of a company losing its competitive standing.

I then go after two important components of ROE, margin and turnover, and limit consideration to companies that are more or less in the top third (top 35% to use a round number) in either operating margin or inventory turnover. Some businesses are, by their very nature, low margin and high turnover (volume), such as a supermarket. Others get high margins but lower volume. I will accept either.

Here’s an interesting issue viz. the retail version of this model. Using inventory turnover there seemed a no-brainer but at restaurants, I thought I might use the more general asset turnover figure. I tested it and it’s OK. But inventory turnover works far better. The question is why, since unlike retailers, these chains don’t have stockpiles of goods sitting around on shelves. (Unless merchandise sales is a meaningful ancillary activity, a restaurant's inventory will often be about 1% or less of total assets, as opposed to 20%-plus often seen at other kinds of non-financial firms.

On reflection, though, I get it. Restaurants inventory is subtle; food and consumables like ingredients, wrappings, napkins, etc. Consumers don't browse through it the way shoppers browse through clothing racks. But the slowdown in the speed at which these very-short-lived stockpiles move can be an indicator that something is wrong. I suppose we might see this as an instance in which little things can have big 2meaning.

Finally, I rank the passing companies and select the top five. (Those in the Portfolio123 community have seen frequent rants of mine regarding portfolios containing so few positions, but the restaurant group is a small one so the sample of five is proportionately reasonable here. Besides, I'm not now building a portfolio per se; I'm looking to establish a stake in a particular theme.) I rank based on a generalized Quality ranking system I often use that considers financial strength, but also return on capital and margin measures that analyze trends over several years and favor trends for the better.

Table 1 shows three sets of backtest results, one set for the past 10 years, one for the past three years and another set just for the last 12 months. I compare the screen-based portfolios to a control group which consists of my entire restaurant universe. The former assumes a 0.25% per trade price penalty as a proxy for transaction costs and that the model is refreshed and the stock lists updated every four weeks.

Table 1

That’s good, but imperfect. The alpha for the latest year shrank because the improved return was accompanied by more volatility. I’d love to avoid that. Realistically, though, I don't think that can happen. I believe we are going to have to get used to more volatility in this industry than we’ve seen in the past as stocks get buffeted this way and that by the flow of news that is often likely to be seen as unfavorable; a new normal. If I can get enough return to make it work, I’ll take it.

The Stocks

Table 2 lists the stocks that currently make the grade under this model.

Table 2

As with the retail screen there are two ways to use a list like this. If I’m comfortable with the results of backtests or simulations and I’m comfortable with the rationale for the model (the latter is important lest one fall into the trap of data mining and investing on the basis of historical results that may be more a matter of good luck than good sense), I invest in all stocks listed and refresh the portfolio at the particular interval (four weeks in this case). I do that a lot. I always understand that this is a probabilities exercise and that some of the names will be clunkers (data tells us a lot, but it can miss things). But it has been my experience that well-conceived well-tested models have produced better overall results than I had been otherwise getting because gems have tended to outnumber dogs.

So, for example, if you decide you hate MCD, so be it. Don’t own it. But at least do yourself a favor and think about the numbers MCD had to have posted to make it into the screen and why they were what they were. Haters feast on MCD and have been doing so for years. But its still here, That has to tell you something.