When Delayed Spotting Meets Delayed Reporting:
Distortions in the Corporate Bond Index Close∗
Andreas C. Rapp† Sean Seunghun Shin‡ Xing (Alex) Zhou§ Qifei Zhu¶
December 2025
Abstract
This paper provides the first systematic evidence that delayed Treasury spot trades
(DTSTs) interacting with delayed trade reporting distort corporate bond pricing at
the index close. Using proprietary MarketAxess data that identify DTSTs and record
spread-agreement timestamps, we show DTSTs execute at stale prices, deviating from
contemporaneous non-DTSTs by 19 bps. These trades mechanically bias end-of-day
close-price estimates, with distortions that intensify with intraday volatility and comove
across bonds. The resulting mispricing passes through to evaluated prices produced
by leading bond index providers. Moreover, delayed reporting of DTSTs impairs price
discovery, both after spread agreement and after the close.
JEL classification: G10, G11, G12, G14, G18, G23.
Keywords: delayed Treasury spotting, delayed reporting, corporate bond market, index close, price distor-
tion, price discovery, investment fund, index provider, benchmark valuation
∗We thank Julien Alexandre and Sinem Uysal at MarketAxess, Katie Gouinlock at Vanguard, and
Damian Park at BlackRock for valuable discussions on institutional details related to delayed Treasury
spotting in corporate bond trading. The analysis and conclusions expressed herein are those of the authors
and do not necessarily reflect those of the Board of Governors of the Federal Reserve System or any other
person associated with the Federal Reserve System.
†Federal Reserve Board of Governors. Email address: andreas.c.rapp@frb.gov
‡KAIST College of Business, KAIST. Email address: sean.shin@kaist.ac.kr
§Cox School of Business, Southern Methodist University. Email address: axzhou@smu.edu
¶NUS Business School, National University of Singapore. Email address: qifeizhu@nus.edu.sg
Distortions in the Corporate Bond Index Close∗
Andreas C. Rapp† Sean Seunghun Shin‡ Xing (Alex) Zhou§ Qifei Zhu¶
December 2025
Abstract
This paper provides the first systematic evidence that delayed Treasury spot trades
(DTSTs) interacting with delayed trade reporting distort corporate bond pricing at
the index close. Using proprietary MarketAxess data that identify DTSTs and record
spread-agreement timestamps, we show DTSTs execute at stale prices, deviating from
contemporaneous non-DTSTs by 19 bps. These trades mechanically bias end-of-day
close-price estimates, with distortions that intensify with intraday volatility and comove
across bonds. The resulting mispricing passes through to evaluated prices produced
by leading bond index providers. Moreover, delayed reporting of DTSTs impairs price
discovery, both after spread agreement and after the close.
JEL classification: G10, G11, G12, G14, G18, G23.
Keywords: delayed Treasury spotting, delayed reporting, corporate bond market, index close, price distor-
tion, price discovery, investment fund, index provider, benchmark valuation
∗We thank Julien Alexandre and Sinem Uysal at MarketAxess, Katie Gouinlock at Vanguard, and
Damian Park at BlackRock for valuable discussions on institutional details related to delayed Treasury
spotting in corporate bond trading. The analysis and conclusions expressed herein are those of the authors
and do not necessarily reflect those of the Board of Governors of the Federal Reserve System or any other
person associated with the Federal Reserve System.
†Federal Reserve Board of Governors. Email address: andreas.c.rapp@frb.gov
‡KAIST College of Business, KAIST. Email address: sean.shin@kaist.ac.kr
§Cox School of Business, Southern Methodist University. Email address: axzhou@smu.edu
¶NUS Business School, National University of Singapore. Email address: qifeizhu@nus.edu.sg
I. Introduction
Index closing prices play a central role in the corporate bond market. They underpin bench-
mark construction, determine mutual fund and ETF net asset values (NAVs), and support
a range of risk-management and oversight applications. Yet producing a reliable “close”
for corporate bonds is inherently difficult. Unlike equities, most bonds trade infrequently
and largely through bilateral, over-the-counter negotiation, so the market often provides few
near-synchronous transactions from which to infer a representative end-of-day price. Al-
though benchmark-tracking mandates and index-based investing have pushed more trading
into index-close windows,1 illiquidity and decentralized trading still complicate the measure-
ment of closing prices.
This paper shows that a specific feature of corporate bond market structure further un-
dermines the reliability of end-of-day prices and can induce systematic mismeasurement. A
common trading convention, delayed Treasury spotting (DTS), interacts with TRACE re-
porting rules so that trades may be recorded at the close even when the economically relevant
component of pricing was determined earlier in the day. In investment-grade (IG) corporate
bonds in particular, market participants often negotiate a transaction during the day as a
spread to a benchmark Treasury, but postpone “spotting” the Treasury (and thus finaliz-
ing the dollar price) until the index close to mitigate benchmark-tracking concerns. Under
current TRACE conventions, these trades are reported only when the benchmark Treasury
is spotted and the dollar price is finalized, and in the public TRACE feed, they are obser-
vationally indistinguishable from trades whose terms were negotiated at the close. Delayed
spotting therefore meets delayed reporting. As a result, closing prints can mechanically mix
stale and contemporaneous credit information while appearing identical in standard data.
Against this backdrop, we ask a simple but fundamental question: To what extent do
these hidden timing frictions distort corporate bond prices, and what are the implications
for benchmark construction and investment fund valuation? We answer this question using
1See Shin, Zhou, and Zhu (2025).
1
Index closing prices play a central role in the corporate bond market. They underpin bench-
mark construction, determine mutual fund and ETF net asset values (NAVs), and support
a range of risk-management and oversight applications. Yet producing a reliable “close”
for corporate bonds is inherently difficult. Unlike equities, most bonds trade infrequently
and largely through bilateral, over-the-counter negotiation, so the market often provides few
near-synchronous transactions from which to infer a representative end-of-day price. Al-
though benchmark-tracking mandates and index-based investing have pushed more trading
into index-close windows,1 illiquidity and decentralized trading still complicate the measure-
ment of closing prices.
This paper shows that a specific feature of corporate bond market structure further un-
dermines the reliability of end-of-day prices and can induce systematic mismeasurement. A
common trading convention, delayed Treasury spotting (DTS), interacts with TRACE re-
porting rules so that trades may be recorded at the close even when the economically relevant
component of pricing was determined earlier in the day. In investment-grade (IG) corporate
bonds in particular, market participants often negotiate a transaction during the day as a
spread to a benchmark Treasury, but postpone “spotting” the Treasury (and thus finaliz-
ing the dollar price) until the index close to mitigate benchmark-tracking concerns. Under
current TRACE conventions, these trades are reported only when the benchmark Treasury
is spotted and the dollar price is finalized, and in the public TRACE feed, they are obser-
vationally indistinguishable from trades whose terms were negotiated at the close. Delayed
spotting therefore meets delayed reporting. As a result, closing prints can mechanically mix
stale and contemporaneous credit information while appearing identical in standard data.
Against this backdrop, we ask a simple but fundamental question: To what extent do
these hidden timing frictions distort corporate bond prices, and what are the implications
for benchmark construction and investment fund valuation? We answer this question using
1See Shin, Zhou, and Zhu (2025).
1
