Working Paper: NBER ID: w30722
Authors: Alexandra M. Tabova; Francis E. Warnock
Abstract: We build the first-ever comprehensive security-level dataset on the size, flows, coupon payments and returns of foreign and U.S. investors’ Treasury portfolios. Private U.S. and private foreign investors hold longer-duration higher-return Treasuries, whereas foreign governments hold shorter-duration lower-return Treasuries, so all else equal private investors should earn higher returns than foreign governments. But all else is not equal. Foreign governments, even with their low-returns low-volatility portfolios, actually earn higher returns because U.S. and foreign private investors’ returns are substantially reduced by poor timing. Moreover, we find that foreigners beat the (non- Fed) market, while U.S. investors earn a small and insignificant 19 basis points less than market returns. From our analysis, no investor-type earns significantly less than the (non-Fed) Treasury market, indicating that differential investors are not the source of the convenience yield. In terms of investor behavior more broadly, our novel dataset allows a direct comparison of the different investors in the Treasury market. Foreign private investors are similar to U.S. private investors and both behave differently and have different preferred habitats than governments.
Keywords: No keywords provided
JEL Codes: F30; G11; G12
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Investor behavior (G41) | Treasury yields (E43) |
| Private investors' portfolio composition (G11) | Returns (Y70) |
| Timing of purchases (G14) | Returns (Y70) |
| US private investors' poor timing (N22) | Reduced returns (G12) |
| Foreign private investors' poor timing (F21) | Reduced returns (G12) |
| Foreign governments' timing (F59) | Returns (Y70) |
| Portfolio composition (G11) | Timing of purchases (G14) |