Liquidity Fades as Treasuries Age

Alain Chaboud, Ellen Correia Golay, Michael J. Fleming, Yesol Huh, Frank M. Keane, and Or Shachar

More than $30 trillion U.S. Treasury debt is outstanding. Less than 4 percent of this amount, which is associated with the most recently issued Treasuries, called on-the-run securities, accounts for 65 percent of average daily trading volume. The remaining portion of the amount outstanding is accounted for by seasoned issues that have been replaced by newer benchmarks, which are referred to as off-the-run securities. In this post, we review the key results in our paper that uses transaction-level Treasury TRACE data to study how trading activity and liquidity evolve as securities move from on-the-run to off-the-run. We show three main patterns. First, off-the-run notes and bonds rely much more on dealer-to-customer intermediation than benchmark securities. Second, trading activity falls sharply and transaction costs increase as securities age. Third, securities that are cheapest to deliver into Treasury futures are an important exception: they trade more actively than other off-the-run bonds of similar age.

Off-the-Run Trading Takes Place in a Different Part of the Market

We rely on Treasury TRACE transactions data to track how trading activity and liquidity change as individual securities season. Such data have been collected since July 2017 and are analyzed in past Liberty Street Economics posts examining the market’s structure, the market for securities that have been announced for auction but not yet issued, and the concentration of trading volume on the last day of each month. In what follows, we mainly focus on 2-year nominal coupon securities, but the results generally carry over to other sectors, including the 5-, 10-, and 30-year nominal coupon securities that are covered in the paper. One limitation is that the data provides us with executed trades, but not quotes or other pre-trade information, and so some of our liquidity measures are biased toward securities that trade sufficiently to allow us to calculate such measures.

The Treasury cash market is split between dealer-to-customer trading and interdealer trading. On-the-run securities dominate the interdealer segment, especially on electronic platforms. Off-the-run securities, by contrast, depend much more on dealer-to-customer activity. From January through June 2024, on-the-run notes and bonds averaged about $262.9 billion per day in alternative trading system (ATS) and interdealer trading, versus $174.8 billion in dealer-to-customer trading (see table below). Off-the-run notes and bonds looked very different: only $51.3 billion per day traded in ATS and interdealer venues, compared with $103.3 billion in dealer-to-customer markets. In other words, when a Treasury goes off-the-run, its trading activity shifts toward a different trading venue, in addition to becoming less active than its benchmark version.

On-the-Run Trading Is Concentrated in the Interdealer Market, While Off-the-Run Notes and Bonds Rely More on Dealer-to-Customer Activity

One reason for this change is that off-the-run securities are harder to match directly. Chaboud et al. (2025) find that only 18 percent of customer trading activity in off-the-run securities had offsetting activity in the same security within the same fifteen-minute interval. This implies that it is difficult to pair buyers and sellers without a dealer taking inventory, which helps explain why off-the-run trading relies more heavily on dealer intermediation.

Trading Volume Declines Once a Security Goes Off-the-Run

The data show that trading volume declines substantially once a security moves off the run. In the 2-year sector, the on-the-run note averages $56.3 billion in daily volume, while the first off-the-run (the second newest issue) averages $5.5 billion and the second off-the-run (the third newest issue) averages $1.6 billion (see charts below). The same pattern appears in other maturities.

The decline in trading frequency is similarly large. The on-the-run, 5-year note averages about 37,900 trades per day, compared with 465 for the first off-the-run. In the 10-year sector, the on-the-run averages about 37,523 trades per day, compared with 661 for the first off-the-run. These differences suggest that the transition to off-the-run status is associated with a meaningful change in how these securities trade.

Changes in the average trade size are also consistent with trading shifting away from high-frequency interdealer platforms and toward dealer-to-customer channels, where larger transactions are more common. In particular, in the 5-year sector, average trade size rises from about $3 million for the on-the-run to $16 million for the first off-the-run. In the 10-year sector, it rises from about $2 million to $10 million, and in the 2-year sector from about $4 million to $11 million.

The results also show that activity in the off-the-run segment is not uniform. Off-the-run securities account for about 30 percent of overall coupon trading volume, with the first and second off-the-runs making up about 15 percent of all coupon activity and deeper off-the-runs accounting for the remaining 15 percent. So, the first few off-the-runs are still an important part of the market, but they nevertheless trade very differently from the securities that replaced them as benchmark securities.

Lower Trading Activity Is Matched by Wider Effective Bid-Ask Spreads as Securities Age

We measure trading costs by calculating effective bid-ask spreads as the difference between the average price of customer buy trades and the average price of customer sell trades divided by the midpoint between the average customer buy and sell prices. Lower effective spreads indicate higher liquidity. We find a clear pattern across maturities: on-the-run securities are the most liquid, first off-the-runs are less liquid, and deeper off-the-runs are less liquid still. In the 2-year sector, the average effective spread rises from 0.66 basis points for the on-the-run to 1.22 basis points for the first off-the-run and 2.23 basis points for the second off-the-run. This pattern is generally similar for the other sectors. In several sectors, the move from on-the-run to first off-the-run roughly doubles the effective bid-ask spread. Thus, these changes are economically meaningful. Reduced liquidity of a security after losing its benchmark status mirrors the decline in its traded quantity.

The liquidity pattern becomes even more pronounced in stressed conditions. In March 2020, for example, the average effective spread for the 2-year on-the-run rose from 0.66 basis points in the full sample to 1.38 basis points. The first off-the-run rose from 1.22 to 4.23 basis points, and the second off-the-run from 2.23 to 7.45 basis points. These widenings suggest that the market’s capacity to absorb selling pressure weakens more for seasoned securities at times of stress.

There is one important caveat when interpreting the results for very deep off-the-runs. Because the effective spread measure is based on executed trades, it can understate illiquidity when securities simply do not trade very often. Some deep off-the-run securities show somewhat narrower effective spreads in March 2020 than in the full sample, but that likely reflects the limits of using transaction-based measures when activity becomes extremely sparse. Infrequent trading is itself often a sign of poor liquidity.

Cheapest-to-Deliver Securities Are an Important Exception

Some off-the-run securities become the cheapest to deliver into Treasury futures contracts. Those bonds matter for hedging, futures delivery, and basis trading, so they should attract more trading interest than otherwise similar off-the-run securities. By regressing daily trading volume at the CUSIP-day level on each bond’s on-the-run/off-the-run status and on a dummy variable that identifies whether a bond is the cheapest-to-deliver (CTD) for the associated Treasury futures contract, we show that being the cheapest to deliver is associated with about $1.31 billion in additional daily trading volume in the 5-year sector. In the 10-year sector, the CTD effect is about an additional $0.75 billion in trading volume, and in the 2-year sector it is about $0.51 billion. The 5-year effect is especially broad-based, showing up across dealer-to-customer, interdealer, and interdealer broker (IDB) activity.

This exception matters because it shows that off-the-run securities do not become less liquid only because they are older. They become less liquid because the uses investors have for them change, the trading venue they rely on changes, and/or the degree of natural two-way trading flow changes.

Summing Up

Market conditions change notably when Treasury securities go off-the-run. Trading shifts away from the interdealer market segment and toward dealer-to-customer intermediation. Volumes and trade frequency fall sharply, while effective bid-ask spreads widen. These patterns appear as soon as a bond loses benchmark status and become more pronounced as the bond seasons further. At the same time, the results for how cheapest-to-deliver status affects trading activity show that this process is not inevitable: when a seasoned bond remains central to hedging and futures delivery, it can retain much stronger trading activity than its age alone would predict.

Alain Chaboud is a former principal economist at the Federal Reserve Board of Governors.

Ellen Correia Golay is a capital markets trading advisor in the Federal Reserve Bank of New York’s Markets Group.

Michael J. Fleming is head of Capital Markets in the Federal Reserve Bank of New York’s Research and Statistics Group. 

Yesol Huh is a former principal economist at the Federal Reserve Board of Governors.

Frank Keane is a former policy and market-monitoring advisor in the Federal Reserve Bank of New York’s Markets Group.

Or Shachar is a financial research advisor in the Federal Reserve Bank of New York’s Research and Statistics Group.

 
@article{AlainChaboud,EllenCorreiaGolay,MichaelJ.Fleming,YesolHuh,FrankM.Keane,andOrShachar2026,
    author={Alain Chaboud, Ellen Correia Golay, Michael J. Fleming, Yesol Huh, Frank M. Keane, and Or Shachar},
    title={Liquidity Fades as Treasuries Age},
    journal={Liberty Street Economics},
    note={Liberty Street Economics Blog},
    number={June 30},
    year={2026},
    url={https://doi.org/10.59576/lse.20260630}
}

You may also be interested in:

Liquidity and Trading Dynamics in the Off-the-Run U.S. Treasury Market

Unlocking the Treasury Market through TRACE

Treasury Market When‑Issued Trading Activity

End‑of‑Month Activity Across the Treasury Market