No. of Recommendations: 3
"I would think extrapolation based on sales (real GDP growth) estimates would be the sensible starting assumption for any firm, then add or subtract based on your solid evidence of how unusually good or bad the firm's prospects are."
I agree that a sensible starting point for extrapolation of earnings could be sales and in turn real GDP growth. The right starting point for sales I believe should actually be slightly lower than real GDP growth, because GDP growth depends on (among other things) capital formation, some of which will happen outside of existing enterprise in which you are measuring growth of sales.
I also believe that there are some important caveats to this general rule. I happen to believe that some part of margin expansion can be more sticky than other parts.
If I build on the paper by mr. Hussman and break down the margin further into its component parts it could look like this:
1. EIBT margins: "like for like" (i.e. assuming same industries)
2. EBIT margins: industry substitution
3. Net margins: interest cost
4. Net margins: tax
Starting from below, 3 and 4 in my view are probably more likely than not to put downward pressure on margins going forward, for all the reasons already mentioned.
On (1) like for like EBIT margins, I believe these are also up over the last couple of decades (sorry I dont' have numbers on hand). There have been many reasons mentioned as why these EBIT margins are up, including government stimulus effects. Whatever the reason, I would probably conclude these should be mean reverting (remember we are talking margins within same industries) and put downward pressure on index margins going forward.
Its only the last part in this breakdown - industry substitution (2) - where i believe the margin COULD be more sticky. If you take out of the index a low margin business and replace it with a high margin business, this will increase the index margin.
Why do I believe this part of margin expansion COULD be more sticky? Simple explanation is that different industries have different margins with the difference being sustained over long periods of time. The more elaborate reason is that I don't believe the economic gods that regulate profitability care so much about margins as they do about returns on capital.
To determine if the high margins in these new businesses are sustainable, it is not enough to look at margins in isolation but we have to look at returns. These are also high - yes - and then we get into the discussion of what denominator to use when calculating returns, accounting capital, true invested capital, replacement capital. Doing the actual adjustments is above my paygrade but I would guess that at least for the last denominator in that list, returns wouldn't look so spectacular anymore.
I am not predicting margins one way or another, just offering my opinion as to the various components of margin expansion and that industry substitution is real, and this part of margin expansion is likely more sticky than others.
