Another massive post. Sorry.
I'm sure most people have seen this, but Ray Dalio estimates that if the FFR hits 4.5%, stocks will drop 20%. This is in line with my base case for the "severe recession" scenario.
https://www.linkedin.com/pulse/starts-inflation-ray-dalio?trk=portfolio_article-card_title4.5% is now very close to the outcome already predicted by futures on the FFR. By July, most market participants are today predicting rates at 4% or above, and about 25% are predicting above 4.5%:
https://www.cmegroup.com/trading/interest-rates/countdown-to-fomc.htmlDalio writes in a folksy way, and does not go into detail on how he did the math, but he appears to be thinking in terms of the Fed Model or a discounted cash flow model. Like myself, Dalio is assuming there will be some risk premium over the FFR for longer duration treasuries, riskier bonds, and especially stocks.
How rising rates influence asset prices:
Interest rates rising relative to inflation causes prices of equities, equity-like markets, and most income-producing assets to go down because of a) the negative effects it has on incomes, b) the need for asset prices to go down to provide competitive returns i.e., “the present value effect”, and c) the fact that there is less money and credit available to buy those investment assets.
Market's prediction vs. Dalio's prediction:
Right now, the markets are discounting inflation over the next 10 years of 2.6 percent in the US. My guesstimate is that it will be around 4.5 percent to 5 percent long term, barring shocks ... In the near term, I expect inflation will fall slightly as past shocks resolve for some items (e.g., energy) and then will trend back up towards 4.5 percent to 5 percent over the medium term. I won't take you through how I arrived at that estimate (which I'm very uncertain about) because that would take too long.
Dalio is willing to conservatively assume a flat yield curve months from now:
If you want to estimate the short rate, decide what you think the yield curve will look like. What's your guesstimate? Mine is that the yield curve will be relatively flat until there is an unacceptable negative effect on the economy. Given my guesstimates about inflation and real yields, I come up with between 4.5 and 6 percent in both long and short rates. However, because I think that the higher end of this range would be intolerably bad for debtors, markets, and the economy, I'm guesstimating that the Fed will be easier than that (though 4.5 percent is probably too easy).
Dalio is being very, very optimistic here for the sake of making a conservative and easy-to-defend estimate. Yield curves tend to un-invert right before the start of recessions, or during the recession.
If another un-inversion is to happen in the near future, either the Fed would have to cut short-term rates OR long term rates would have to rise. Recent Fed statements have been about the need to keep rates higher for a long period of time to ensure the inflation fire is fully extinguished - unlike what was done in the 70's. So if we disregard the possibility of any rate cuts next year no matter how bad it gets, then the long-duration end of the curve will have to rise and an un-inversion might look like a 4.5% FFR and maybe a 6% 10-year yield. Stocks then, would have an earnings yield higher than 6%, which means the PE of the S&P500 would have to be significantly less than (1/6=) 16.7. That is, at PE=16.7 you would be getting paid zero risk premium to be in stocks versus safer 10y treasuries, so you wouldn't buy or hold stocks at such a high price.
Actually, the yield curve is not even fully flat right now. The overnight rate is 2.25%-2.5%, the 1 month rate is 2.68%, and the 10y is 3.45%. The 10y/2y spread is now more inverted than anytime since right before the 2000 recession/bear market
and there's nonetheless a ~1% spread between overnight and 10y rates. If both Dalio and the futures market are predicting a 4.5% FFR, they should predict at least a 5.5% 10-year yield, and that's probably an understatement.
These are more arguments in favor of the S&P500 soon having a PE ratio in the mid-teens instead of near 20.
If you factor in dramatically reduced earnings forecasts amid a recession, then the 20% drop we're talking about could become much bigger. This is the part we cannot know. Valuation metrics go out the window as soon as a recession starts, because there's no way to tell in advance how far earnings will fall, or how long they will take to recover. As earnings fall, companies look deceptively more expensive according to earnings-based metrics.
Suppose the following happens in June 2023:
FFR = 4.5%
10y yield = 5.5% (same premium as today)
S&P 500 expected earnings yield: 7% (1.5% risk premium over 10y treasury)
S&P 500 forward P/E: 14.3 (a 28.5% fall from last week's PE of 20)*
Recession starts but the market doesn't realize it yet.
*Note: Based on Yardeni's forward PE, forecasters are expecting about a ((20/16.9)-1=) 18% earnings increase in the next 12 months, assuming no recession. If, let's say, 12% of that happens before the recession, then maybe the nominal price of stocks only falls (28.5% fall in PE minus 12% increase in E =) 16.5% between now and then.
And then the following happens by December 2023:
Recession should be a headline by now
FFR = 4.5%
10y yield = 5.5%
S&P 500 forward earnings drop 30%**
S&P 500 risk premium over treasuries increases from 1.5% to 2% due to uncertainty about the depth of recession
S&P 500 expected earnings yield: 7.5% (10y yield + 2% risk premium)
S&P 500 forward PE: 1/7.5 = 13.3***
**E.g. If earnings go up 12%, then the forecast drops 30% from there, that'd be down to 78.4% of today's earnings. -30% is a guess. Compare that guess to -50.6% in 2001 or -77.5% in 2008 and it looks optimistic.
***The trailing PE ratio normally goes higher during recessions as earnings fall, but investors look beyond the recession and price a volatile series of forward estimates. So we can't just multiply this formula-driven PE by our earnings guess and arrive at a call of where the bottom will be. In general, Yardeni's data show when the S&P500 forward PE is in the low-to-mid teens, that's a good time to buy.
In the scenario described above, S&P500 forward earnings would be (ttm earnings of 203.88 * 78.4% =) 159.84, a level similar to the forward estimates in 2017 when the S&P500 was about 2700 (29% lower than today's prices, the PE was around 25, and the 10y treasury was around 2.5%). However, investors would be unwilling to buy a stock market forward earnings yield that is not significantly higher than the 10y treasury yield. Treasuries will look attractive as inflation expectations and earnings forecasts both collapse amid the recession. IG bonds yielding even more than the treasuries - perhaps 6% to 8% - might be even more attractive. What kind of WR could those kind of yields support? [wipes drool away]