No. of Recommendations: 19
I thought I recalled an analysis by Jim of forward returns versus Price/Sales bucket for DG but can't find it using datahelper.
I'm not sure, but I don't think I did that for DG.
I have done it for a few other firms that price/sales has some value, like GOOGL.
Price-to-sales makes some sense for DG, *provided* you are comfortable with the built in assumptions that (1) net margins will go back to their old normal range give or take, and of course (2) that valuation multiples will at least visit their old normal levels. Frankly I'm not 100% comfortable with either of those.
The problem with such an analysis for DG is that P/S was in one range for a while, then a different higher range. The reason is simple: like Berkshire, DG pays a high effective tax rate, so the tax cuts at end 2017 made a big difference to profitability. The tax rate averaged 36.5% 2013-2017, then averaged 21.9% 2018-2022. Consequently net margins went up, from an average of 5.62% to an average of 6.74%. Each dollar of revenue started generating 20% more net profit. Other things being equal, if you assume that the tax cuts are permanent, the "normal" price-to-sales ratio should rise by the same 20% amount. In fact the average multiple rose more than that: the average went from around P/S 1.0 to P/S 1.41 in the last five years. (very slightly different time slices, but you get the idea). Perhaps some of that was market exuberance.
So, several moving parts. One way to come at it: Let's take the older number of 1.0 times sales as the somewhat more conservative number for normal, at that old net margin. Let's conservatively assume that the tax cuts aren't permanent. Headline went from 35% to 21%, I have been pencilling in a moderate rebound to about 25%. Roughly speaking, I'm assuming that 70% of the tax cut is permanent and 30% temporary. So, ignoring any other moving parts, I should expect the new normal P/S ought to rise from 1.0 to 1.14. (being 70% of the 20% bump from tax cuts).
Because of all this jiggery-pokery, rather than looking at the historical relationship between P/S and forward returns, I think it's better to simply estimate what the trajectory of sales and price will be in future from these figures. Trailing four quarters sales are $174, so at P/S 1.14 that means that the price "should" be about $198 now on a cyclically adjusted basis. Then add whatever you think the sales growth rate will be at the end of your forecast horizon. Sales per share have risen 12.5%/year in the last dozen years, so pick a forward number for math purposes, say 8%. That suggests that sales would rise 36% to $237 in the next four years, which at 1.14 times sales would be a (nominal) price target of $270, or 24%/year compounded from today's $113.66 price, minus inflation, plus dividend yield.
Modify assumptions to suit your greed and fear : )
Jim
