The small-cap blues | Financial Times

Good morning. The National Bureau of Economic Research officially declared yesterday that US long-term interest rates have “gone bananas”. Actually this is not true, but it should be. The 10-year Treasury yield has ripped through 4.6 per cent. This is getting a little scary, but looking on the bright side, the curve is less inverted now? Email us: robert.armstrong@ft.com and ethan.wu@ft.com.

Why small caps have not worked

Back in July, we wrote about the case for small caps. It came down to two factors: valuations and growth. Multiples looked reasonable, especially compared with the expensive S&P 500. The Russell 2000’s price/book ratio was well below the historical average. And the soft landing story was starting to come together. As recession fears got priced out, the Russell would rally, we thought.

Well, er, about that:

The S&P has been falling since July, but the Russell has fallen further. On valuations, the Russell has gone from cheap to ultra-cheap. Its price/book multiple is in the bottom quintile of its range stretching back to 1995. Generically, small caps tend to be more volatile in both directions. But soft landing is still many people’s base case and small caps remain far cheaper than the S&P. What derailed the case for small caps?

One thing is a downward revision to market growth expectations. Soft landing, we should remember, still implies a slowdown. Ryan Hammond of Goldman Sachs reckons that market pricing of GDP growth has fallen in the past several months from something like 3 per cent to 2 per cent, pointing to measures like defensives’ outperformance relative to cyclicals.

Another is the ambiguous spot we’re at in the economic cycle. As a rule, early cycle is when you want to own the Russell. Small caps, with their heavy representation of unprofitable minnow companies, tend to sell off hard as the cycle matures or falls into recession, often followed by a mighty rally once growth picks up again. We showed you this chart back in July:

But lately, it’s not been quite so clear if we’re in early cycle, late cycle or some ambiguous spot that defies the cycle framework. As such, it’s not clear small caps are due for an early-cycle rally anytime soon.

Lastly, higher interest rates are biting for small caps. As we wrote elsewhere in the Financial Times, S&P 600 interest expense per share has hit a record. Small-cap indices are laden with thin-margin companies, and plenty of unprofitable ones, so it’s not a stretch to figure the probability of defaults must be rising:

One point of speculation. The rise of private equity in the past few decades may have sapped some vitality from the small-cap universe. As capital has flooded into PE funds, it makes sense that they would have picked through the small-cap universe for companies with potential, leaving a weakened group behind. That would fit with a striking fact about the Russell: the share of unprofitable companies (on a 12-month forward basis) is one in three.

Still, any asset class, however junky, makes sense at the right price. With small caps historically very cheap, perhaps the buy case has gotten stronger. If you’ve a view, let us know. (Ethan Wu)

Rates and stocks: trying to kill a zombie with maths

On Tuesday, I argued, not for the first time, against the zombie idea that rising interest rates are especially bad for growth stocks. To summarise: when inflation expectations and interest rates rise, it’s not just the discount rate on future profits that changes. Growth rates change too, for example. So the “rates up = growth underperforms” theory is at best a wild oversimplification and at worst just false.

But analysts and pundits keep at it. As if to taunt me, Bloomberg published this in a market wrap this week:

The threat of tight policy is undoing some of the market’s biggest gains this year in the high-flying tech stocks. These growth companies are prized for their long-term prospects but hold less appeal when future profits get discounted at higher rates.

But rates rose like crazy right through the summer, and tech stocks did great! What was happening then? Argh!

Yet another final attempt, then, to put an end to this intellectual virus. Let’s try it with numbers this time. Below is a very simple net present value analysis of two imaginary companies, one growth-y, one value-y. Sorry to those of you reading on mobile, who will have to squint:

Both companies have the same profits as of year one, but one will grow at 5 per cent for the next 10 years, the other at 2.5 per cent (I know, growth/tech companies usually grow faster than 5 per cent, but I wanted a flat rate over 10 years so I shaded it down). I use the 30-year Treasury yield for the risk-free rate, specifically the one from December 2021, when stocks peaked. I borrowed the equity risk premium calculated by Aswath Damodaran of NYU, again from the end of 2021. To calculate the terminal value of each stock at the end of 10 years, I shaved a percentage point from both growth rates. Sorry for all this boring arithmetic, I will make an interesting point soon, I promise.

As you can see from the net present values at lower left — the prices of the two stocks, as it were — a lowish discount rate and a highish profit growth rate combine to make the growth stock look very expensive, at 38 times this year’s earnings. Most of that value resides more than 10 years in the future, in the terminal rate. The value stock is less expensive, at 20 times earnings, and its value is weighted more towards the present. 

Now let’s suppose there was a big inflation shock and rates rose or, to put in another way, let’s use the 30-year Treasury rate and the ERP from today, not late 2021, while keeping everything else the same:

The prices of both stocks get crushed by the higher discount rate, but the growth stock, because more of its value is out in the future, gets the worst of it. Its value falls by half, versus a third for the value stock. 

Now let’s make another assumption, to make the analysis more closely resemble the real world: the two companies have inflation-offsetting pricing power, but to different degrees. Let’s suppose the growth company has higher pricing power, and can increase its nominal prices enough to move its profit growth rate to 7.5 per cent a year, while the value company can push to 4 per cent a year. Let’s push the terminal growth rates up a bit, too. Now the picture is very different:

Now the value stock falls by more than the growth stock. Of course a little fiddling with the inputs would get you a different result, but that is exactly the point. Once you add in just one more variable — pricing power — the “rates up growth underperforms” theory goes to pieces. 

And it is not arbitrary to assign more pricing power to the growth stocks. Part of what defines many growth stocks is that their businesses have high barriers to entry. Think of companies most people are talking about when they are talking about tech and rising rates: the magnificent seven. Apple and Microsoft are defined by pricing power; Meta and Google perhaps less so, but as advertising businesses go, they must have more leverage with clients than anyone else; if you want an AI chip stack, it’s Nvidia or no one right now; Amazon and Tesla have niches where they dominate, too. This is part of why we’ve argued in the past that Big Tech may prove surprisingly defensive in a downturn.

None of this is to deny that the prices of growthy tech stocks sometimes seem to have an inverse relationship with Treasury yields. The problem is we don’t seem to have a good theory of why this happens when it does. I suspect the theory will have to pull from the arsenal of behavioural finance, with its emphasis on bias and overreaction, rather than adding more macroeconomic variables. More on that tomorrow. 

One good read

People on X née Twitter are angry about this New Yorker story about the Bankman-Fried family, saying it lets Sam’s academic haute bourgeois parents off the hook too easily, given their own links to FTX. But let me stick up for the writer, Sheelah Kolhatkar: maybe she trusted her readers to do the putting-on-the-hook themselves?