Two Data items today:
1: US Treasury debt issuance and yields. We’ve heard concerns from clients that the Treasury will have to ramp up issuance over the rest of 2021 and this will cause a backup in yields. We are sharpening our pencils on this issue and will report back next week. As we wrote last night, our base case has 10-year Treasury yields rising to +1.50 percent in fairly short order and that should be helpful for cyclical stocks. Still, supply-demand imbalances could make that move sharper than markets would welcome.
This line of questioning did get us to wondering about the long-term relationship between Treasury issuance and 10-year yields, so we pulled the data back to 1980, as shown in the chart below. The black line shows the annual growth rate in the dollar amount of Treasuries held by the public (i.e., not those held by Social Security and other government programs). The blue shows 10-year Treasury yields.
The relationship from 1980 – 2000 is clear: lower growth rates in Treasuries outstanding correlate with lower yields. The trend to less Treasury issuance peaked in 2000, when the amount of Treasuries in circulation actually shrank because the US ran a budget surplus in that year. Readers with long memories will recall that Treasury stopped issuing 30-year paper in February 2002 because it decided it did not need the money and shorter term rates were more attractive. That decision did not age well. Treasury went back to issuing 30-years in 2006.
Since 2000, there is simply no tie between incremental Treasury issuance and rates. The graph tells the story. The dollar amount of Treasuries outstanding has increased by 5-20 percent every year (black line) and rates (blue line) have plumbed ever-lower depths.
Takeaway: the correlation between incremental Treasury issuance and yields ended 20 years ago. Why? US money velocity (GDP/M2 money supply) peaked 2 decades ago, as we’ve recently discussed, going from a high of 2.2x in 1997 to 1.1x today. The money created by Treasury issuance post-2000 has not had the same impact on economic growth, and therefore inflation, as before that year. Lower levels of inflation mean lower Treasury yields, which explains the lack of a post-2000 relationship between issuance and yields.
#2: Dividend yields around the world, present and as compared to 2019 payouts, as a measure of the market’s confidence in an eventual recovery in corporate earnings. For example, the S&P 500 has a 4-quarter trailing yield of 1.3 percent using payouts from the IVV ETF as a proxy for those payments. Using 2019’s actual IVV payouts, the S&P’s yield is 1.4 percent, 8 percent higher. This signals the market’s confidence (entirely correct, given the last 2 quarters of reported earnings) that payouts will soon equal 2019 levels.
Here is the same math for a few global equity indices:
- MSCI All-World Index (ACWI):
- 1.4 percent yield based on trailing 12-month payouts
- 1.8 pct yield based on 2019 payouts
- 2019 – 2021 Gap: 29 percent
- MSCI All World ex US (ACWX):
- 2.0 percent yield based on trailing 12-month payouts
- 2.8 pct yield based on 2019 payouts
- 2019 – 2021 Gap: 40 percent
- MSCI EAFE (non-US developed economies, EFA):
- 2.3 percent yield based on trailing 12-month payouts
- 2.7 pct based on 2019 payouts
- 2019 – 2021 Gap: 17 percent
- MSCI Emerging Markets (EEM):
- 1.5 percent yield based on trailing 12-month payouts
- 2.4 percent based on 2019 payouts
- 2019 – 2021 Gap: 60 percent
Takeaway: markets have the greatest confidence – by far – that US large cap companies will not just recoup their pre-pandemic earnings power but soon reinstate dividend payouts that confirm this fact. The S&P 2019 – 2021 dividend yield gap is just 8 percent, while the EAFE gap is 17 percent and the Emerging Markets gap is 60 percent. While we like MSCI Europe (IEUR, 19 percent gap), the EM gap is not in our opinion a market anomaly. It rightly represents the reduced earnings power from Chinese companies, as we have been discussing over the last few weeks.