This blog post is the first of a series on fixed income markets that the Department of the Treasury will release in the coming weeks to share our perspective on the available data, discuss key structural and cyclical trends, and reiterate our policy priorities. We begin with a summary of liquidity conditions in the U.S. Treasury market.
The U.S. Treasury market is the deepest and most liquid government securities market in the world. Treasuries play a unique role in the global economy, serving as the primary means of financing the U.S. federal government, a critical store of value and hedging vehicle for global investors and savers, the key risk-free benchmark for other financial instruments, and an important conduit for the Federal Reserve’s implementation of monetary policy.
Like nearly every other financial market, the Treasury market is in a state of transition. The structure of the Treasury market has evolved significantly over the past two decades, as described in the Joint Staff Report (JSR) on the events of October 15, 2014. Advancements in technology, and the associated growth in high-speed electronic trading, have contributed to changes in intermediation and the provision of liquidity in the Treasury market. Most notably, principal trading firms (PTFs) are increasingly prevalent and now account for the majority of trading and standing quotes in the order book in both futures and the inter-dealer cash market. By contrast, bank dealers still account for a majority of secondary cash market trading overall, but they comprise well under half of the trading and quoting activity in the inter-dealer cash market.
This post starts with a review of several traditional measures used to assess the state of liquidity in the Treasury market, some of which were discussed in the recent blog series published by the Federal Reserve Bank of New York, which has long acted as Treasury’s fiscal agent. We also review additional measures to test whether traditional metrics may, on average, fail to capture liquidity conditions in seasoned securities or abrupt changes in liquidity conditions.
While no individual metric is dispositive, these measures together suggest that liquidity in the Treasury market is consistent with historical levels. Importantly, however, these data only reflect a portion of transactions in the inter-dealer market—electronic platforms where traditional dealers interact with each other and, increasingly, with PTFs. Official sector access to data related to transactions in the dealer-to-customer market remains limited, which is the one of the reasons why Treasury solicited feedback in the recent Request for Information on how to improve official sector access to Treasury market data.
Average Daily Trading Volumes: Trading volumes are often viewed as a proxy for liquidity because high volumes suggest that buyers and sellers are able to regularly meet and transact. As shown below, Treasury trading volumes are well within recent averages. Of course, while volumes are generally higher in liquid markets, it is worth noting that volumes may also be elevated in periods of increased volatility where liquidity is less robust.
Bid/Ask Spread: The spread between the best bid and offer prices for Treasury securities illustrates the cost of transacting in a “typical” size. Bid-ask spreads have remained tight over the last five years, indicating that Treasury investors should be able to buy and sell securities at nearly the same price. However, it is important to note that bid-ask spreads also remained very tight in both futures and the inter-dealer cash market throughout the day on October 15, 2014, a period of significant volatility in which few would characterize Treasury market liquidity as robust.
Trade Size: Average trade sizes for most Treasury securities with longer tenors have been stable. Starting in 2013, trade sizes for 2-year Treasury securities, the most sensitive to the short-term interest rate outlook, began to decline. This decline coincided with the Fed signaling it would taper its bond purchases, which likely led to relatively less market certainty regarding the path of monetary policy. Today, 2-year trade sizes are close to the same levels as other Treasury maturities.
As with the prior metrics, it is important to note that average trade sizes, as a standalone measure, are an incomplete measure of liquidity. To the extent that electronic trading platforms and methods are breaking up larger trades into a series of smaller transactions, a decline in average trade sizes may reflect changes to market structure rather than a deterioration in liquidity.
Price Impact of Trades: The estimated price impact per $100 million of net order flow has generally returned to levels prevailing prior to 2008. However, the 10-year price impact measure has edged higher since the beginning of 2015, which may reflect rising volatility in trading conditions from historically low levels alongside broader uncertainties in global markets.
Market Depth: Market depth— the size of the best bids and offers in the order book—has declined somewhat since early 2013, which may have implications for the ability of market participants to trade large blocks of Treasuries. However, this trend is most notable in 2-year securities and, similar to average trade size, peaked during 2011-2012 when short-term interest rate expectations were exceptionally stable. Moreover, current market depth is near average levels observed over a longer historical period.
Liquidity Index: As noted above, there is no definitive measure of liquidity. Constructing a composite index that incorporates several traditional measures of liquidity may provide a more comprehensive picture. The index shown below includes several characteristics of the order book, such as bid-ask spread, top-of-book prices, and market depth, to quantify the ease with which investors can transact. As reflected by this index, current liquidity conditions are broadly similar to levels that have prevailed since 2010. Additionally, as reflected by the red bars in the charts below, which depict liquidity conditions on days when FOMC statements are issued and non-farm payroll data is released, Treasury market liquidity appears to diminish most around the release of market-moving economic news. Outside of these events, liquidity appears largely stable.
In summary, we find little compelling evidence of a broad-based deterioration of Treasury market liquidity using traditional metrics. However, as noted earlier, these traditional metrics are based on a partial view of trading conditions in the cash market—namely trading in on-the-run securities in the inter-dealer market—and thus do not account for trading conditions in off-the-runs, which are traditionally less liquid. Moreover, while traditional measures may be indicative of general market conditions, they may not capture other changes in liquidity conditions.
G-Spread: Some market participants have noted that liquidity in the Treasury market has diverged between more-liquid “on-the-run” securities (i.e., the most-recently-issued securities of a given tenor) and less-liquid “off-the-run” securities. One measure of this dynamic, which we refer to as the “G-Spread,” is the additional yield that a seasoned Treasury security pays over the yield on a recently-issued security of the same tenor. If one attributes this spread solely to the difference in liquidity between these two securities, then a widening yield spread across tenors may reflect a deterioration in liquidity. What we find is that G-spreads have increased slightly since 2013 but remain a fraction of levels observed in 2009.
Changes in Liquidity Index: Another point made by market participants, and a question posed by the JSR, is whether the Treasury market is becoming more vulnerable to periodic episodes of intraday volatility and an associated deterioration in liquidity as high-speed, algorithmic trading becomes increasingly prevalent.
Changes in the composite index described above can be viewed as a proxy for how frequently liquidity conditions change over time. As shown in the chart below, the volatility of this index is consistent with what we have observed over the last six years. Moreover, while we do observe episodic reductions in liquidity around the release of market-moving economic news such as non-farm payrolls or FOMC decisions, it appears that the extent of the reductions are fairly consistent over time.
In light of these metrics, both traditional and new, our strongest conviction is that the Treasury market—much like other secondary markets—is in a period of transition, driven by structural and cyclical factors. The Request for Information (RFI) issued by Treasury in January asked several questions about the evolution of the Treasury market and the implications for liquidity. The comment period for the RFI closed on April 22, and we look forward to updating our analysis with information provided in response to the RFI.
James Clark is the Deputy Assistant Secretary for Federal Finance, and Gabriel Mann is a policy advisor in the Office of Debt Management at the U.S. Treasury Department.