Search

We model the joint dynamics of intraday liquidity, volume, and volatility in the U.S. Treasury market, especially through the 2007–09 financial crisis and around important economic announcements. Using various specifications based on Bauwens and Giot (2000)’s Log-ACD(1,1) model, we find that liquidity, volume, and volatility are highly persistent, with volatility having a lower short-term persistence than the other two. Market liquidity and volume are important to explaining volatility dynamics but not vice versa. In addition, market dynamics change during the financial crisis, with all variables exhibiting increased responsiveness to their most recent realizations. Our models also reveal different market dynamics around announcements. Finally, we introduce new measures of liquidity risk that are useful for continually monitoring liquidity conditions and the risk of liquidity stress in the market.

Market disruptions in response to the COVID pandemic spurred calls for the consideration of marketwide central clearing of Treasury securities, which might better enable dealers to intermediate large customer trading flows. We assess the netting efficiencies of increased central clearing using nonpublic Treasury TRACE transactions data. We find that central clearing of all outright trades would have lowered dealers’ daily gross settlement obligations by roughly $330 billion (60 percent) in the weeks preceding and following the market disruptions of March 2020, but nearly $800 billion (70 percent) when trading was at its highest. We also find that expanded central clearing would have substantially lowered settlement fails. The estimated benefits would likely be greater if dealers’ auction purchases were included in the analysis or if the increased central clearing included repo transactions.

We study the workup protocol, an important size discovery mechanism in the U.S. Treasury market. We find that workup order flow shocks explain 6%–8% of the variation of returns on benchmark notes and, across maturities, 10% of the variation of the yield curve level factor. Information related to proprietary client order flow is more likely to show up in workup trades, whereas information derived from public announcements tends to come through preworkup trades. Our findings highlight how the nature of information affects the trade-off between speed and execution price when informed traders choose between the lit and workup channels. Received May 3, 2017; Editorial decision August 1, 2018 by Editor Thierry Foucault. Authors have furnished an Internet Appendix, which is available on the Oxford University Press Web site next to the link to the final published paper online. Internet Appendix tables are numbered with “IA” prefix.

A series of fourteen 2-aryl-3-phenyl-2,3-dihydro-4H-pyrido[3,2-e][1,3]thiazin-4-ones was prepared at room temperature by T3P-mediated cyclization of N-phenyl-C-aryl imines with thionicotinic acid, two difficult substrates. The reactions were operationally simple, did not require specialized equipment or anhydrous solvents, could be performed as either two or three component reactions, and gave moderate–good yields as high as 63%. This provides ready access to N-phenyl compounds in this family, which have been generally difficult to prepare. As part of the study, the first crystal structure of neutral thionicotinic acid is also reported, and showed the molecule to be in the form of the thione tautomer. Additionally, the synthesized compounds were tested against T. brucei, the causative agent of Human African Sleeping Sickness. Screening at 50 µM concentration showed that five of the compounds strongly inhibited growth and killed parasites. Published version

This article examines market liquidity in the postcrisis era in light of concerns that regulatory changes might have reduced dealers’ ability and willingness to make markets. We begin with a discussion of the broader trading environment, including an overview of regulations and their potential effects on dealer balance sheets and market making, but also considering additional drivers of market liquidity. We document a stagnation of dealer balance sheets after the financial crisis of 2007–2009, which occurred concurrently with dealer balance sheet deleveraging. However, using high-frequency trade and quote data for US Treasuries and corporate bonds, we find only limited evidence of a deterioration in market liquidity.

We examine the financial conditions of dealers that participated in two of the Federal Reserve's lender-of-last-resort (LOLR) facilities—the Term Securities Lending Facility (TSLF) and the Primary Dealer Credit Facility (PDCF)—that provided liquidity against a range of assets during 2008–2009. Dealers with lower equity returns and greater leverage prior to borrowing from the facilities were more likely to participate in the programs, borrow more, and, in the case of the TSLF, at higher bidding rates. Dealers with less liquid collateral on their balance sheets before the facilities were introduced also tended to borrow more. The results suggest that both financial performance and balance sheet liquidity play a role in LOLR utilization.

This paper studies the effects of a recent tick size reduction in the U.S. Treasury securities market. We find significantly narrower bid-ask spreads, increased trading activity, improveddepth within one old tick despite lower overall market depth, and improved price efficiency.Moreover, slow traders become more competitive in liquidity provision and price improvement.Finally, high-frequency price discovery shifts to the smaller-tick cash market from the associatedfutures market, supporting the idea that a finer pricing grid allows traders to act promptly on even small information signals otherwise not profitable. Overall, we conclude that the tick sizereduction improves market quality.

The U.S. Treasury market is highly intermediated by non-bank principal trading firms (PTFs). Limited capital forces PTFs to end the trading day close to flat. We construct a continuous time market making model to analyze the trade-off faced by a profit maximizing firm with overnight inventory costs, and develop closed-form representations of the optimal price policy functions. Our model reveals that bid-ask spreads widen as the end of the trading day approaches, and that increases in order arrival rates do not always lead to higher price volatility. Our empirical analysis shows that Treasury security trading costs increase as the close of trading approaches, consistent with model predictions.

Following an earlier joint FEDS Note and Liberty Street Economics blog post that examined aggregate trading volume in the Treasury cash market across venues, this post looks at volume across security type, seasoned-ness (time since issuance), and maturity. The analysis, which again relies on transactions recorded in the Financial Industry Regulatory Authority's (FINRA) Trade Reporting and Compliance Engine (TRACE), sheds light on perceptions that some Treasury securities—in particular those that are off-the-run—may not trade very actively. We confirm that most trading volume is made up of on-the-run securities, especially in venues where the market has become more automated. However, we also find that daily average volume in off-the-run securities is still a meaningful $157 billion (27 percent of overall volume), and accounts for a large share (41 percent) of trading in the dealer-to-client venue of the market.

This joint FEDS Note and Liberty Street Economics blog post from staff at the Board of Governors and Federal Reserve Bank of New York aims to share initial insights on the Treasury cash transactions data reported to Financial Industry Regulatory Authority (FINRA)'s Trade Reporting and Compliance Engine (TRACE).

The Term Securities Lending Facility (TSLF) was recently introduced by the Federal Reserve to promote liquidity in the financing markets for Treasury and other collateral. We evaluate one aspect of the program – the extent to which it has narrowed repo spreads between Treasury collateral and lower quality collateral. We find that TSLF operations have precipitated a significant narrowing of repo spreads. More refined tests indicate the market conditions and types of operations associated with the program's effectiveness. Various additional tests, including a split sample test, suggest that our findings are robust.

We assess dealer behavior in the specials market for US Treasury securities by comparing dealer participation in the Federal Reserve's securities loan auctions with prices in the private market. Dealer behavior is generally consistent with the law of one price and apparent violations can largely be explained by institutional differences between the private market and the Fed's program. However, for auctions that are effectively noncompetitive, dealers regularly pass up true arbitrage opportunities and frequently overpay to borrow securities. Dealers apparently do not realize that certain auctions are noncompetitive, even though the information needed to discern this fact is publicly available in advance.

We examine the financial conditions of dealers that participated in two of the Federal Reserve’s lender-of-last-resort (LOLR) facilities -- the Term Securities Lending Facility (TSLF) and the Primary Dealer Credit Facility (PDCF) -- that provided liquidity against a range of assets during 2008-09. Dealers with lower equity returns and greater leverage prior to borrowing from the facilities were more likely to participate in the programs, borrow more, and--in the case of the TSLF -- at higher bidding rates. Dealers with less liquid collateral on their balance sheets before the facilities were introduced also tended to borrow more. There also appear to be some interaction effects between financial performance and balance sheet liquidity in explaining dealer behavior. The results suggest that both financial performance and balance sheet liquidity play a role in LOLR utilization.

THE UNITED STATES achieved its third consecutive federal budget surplus--a record $237 billion--in fiscal 2000.(1) This string of surpluses has allowed the Treasury Department to begin to pay off the national debt. After more than tripling from the early 1980s to the mid-1990s, outstanding marketable U.S. Treasury securities fell from just under $3.5 trillion in March 1998 to $3.0 trillion in July 2000. The Office of Management and Budget (OMB) projects that the surpluses will continue, causing the debt held by the public to be fully redeemed by 2012.(2) Although a remarkable achievement, the paydown of the debt also raises some concerns. U.S. Treasury securities play a central role in the implementation of monetary policy and in the efficient working of financial markets more broadly. By reducing and possibly someday eliminating the stock of these securities, the debt paydown raises questions about how monetary policy will be conducted in the future and how financial markets will adapt to the diminished supply of this key instrument. The Treasury's introduction of a debt buyback program in January 2000 and the striking inversion of the Treasury yield curve this year have heightened interest in these questions and spurred discussion as to which assets might be suitable Treasury substitutes. Treasury securities play several critical roles in financial markets. Because these securities are considered free from default risk and are highly liquid across a wide range of issues, their yields are used as a proxy for risk-free interest rates. These properties, together with the presence of well-developed derivatives markets in which investors can sell Treasuries short, make them a useful reference benchmark and hedging instrument for other fixed-income securities. Their creditworthiness and liquidity also make Treasury securities a popular reserve asset for numerous financial institutions and the primary asset of the Federal Reserve. Some of the very features that make Treasury securities an attractive benchmark and reserve asset are likely to be adversely affected by the debt paydown. In fact, recent events suggest that the reduced supply of Treasuries may already be disrupting the market and that more such disruptions may be in the offing. In February 2000, for example, the Treasury announced that its one-year bill would henceforth be issued only every thirteen weeks rather than every four weeks. As the last bill auctioned on the old cycle aged, the bill became very expensive to borrow in the market for repurchase agreements (repos). On May 31, for instance, dealers had to lend out funds at a very low 2.25 percent annual rate in order to secure the one-year bill as collateral. The liquidity of the issue in the cash market also suffered, with bid-ask spreads widening and trading volume plunging. At the same time, the issue became extremely expensive relative to other Treasuries of similar maturity. With the debt paydown under way, market participants are already moving away from Treasury securities as a reference benchmark and hedging device and toward the debt securities of government-sponsored enterprises (such as Fannie Mae) and state-chartered corporations, and toward interest rate swaps.(3) These other instruments are liquid (although not as liquid as Treasuries), the debt securities can be borrowed in reasonably active repo markets, and a futures market was recently introduced for agency securities (and is being discussed for corporate securities). Furthermore, the credit risk in these instruments actually makes them potentially better hedging vehicles than Treasuries, because it can result in them trading at prices that track more closely those of other fixed-income securities that also have credit risk. Agency securities and swaps, in particular, are increasingly used to hedge positions, price new securities, and evaluate existing securities in U.S. fixed-income markets. The Federal Reserve System has meanwhile taken several measures to adapt its conduct of monetary policy to the debt paydown. …

This article takes a close look at a year in the U.S. Treasury securities market and tries to explain the sharpest price changes. The authors attribute each of the twenty-five largest price shocks during August of 1993 to August 1994 period to a just-released macro-economic announcement. They also measure the bond market9s average reaction to various announcements and the reactions to surprise in the announcements headline numbers. The strongest responses are found to come from the employment, producer price index, and fed funds target rate announcements. Significant responses to U.S. Treasury security auctions results are documented for the first time.

This paper examines how workers use 401(k) plans by examining their participation, contribution, and withdrawal decisions. Sixty-five percent of eligible workers participate in 401(k) plans. Employee participation rises with income, age, job tenure, and education. While participation also rises if the employer matches contributions, 401(k) participation does not grow with the rate of matching. When pension plan assets are withdrawn in lump-sum distributions before retirement, just 28 percent of distribution recipients (representing 56 percent of distribution assets) roll over the withdrawn funds into tax-qualified savings plans. Our findings suggest that many workers, particularly those with low incomes, do not use 401(k) plans to save for retirement.

Long-term interest rates hit record-low levels in 2012 but have since increased substantially.

Although China now has one of the largest government bond markets in the world, the market has received relatively little attention and analysis. We describe the history and structure of the market and assess its functioning. We find that trading in individual bonds was historically sparse but has increased markedly in recent years. We find also that certain announcements of macroeconomic news, such as China’s producer price index (PPI) and manufacturing purchasing managers’ index (PMI), have significant effects on yields, even when such yields are measured at a daily level. Despite the increased activity in the market, we are able to reject the null hypothesis of market efficiency under two different tests for four of the most actively traded bonds.

The issues of liquidity and price transparency in derivatives markets have taken on greater import given regulatory efforts under way to improve their transparency. To date, the lack of transaction data has impeded the understanding of how the inflation swap and other derivatives markets operate. This article broadens that understanding by using a novel transaction data set to examine trading activity and price transparency in the quickly growing U.S. inflation swap market. The authors find that the market appears reasonably liquid and transparent, despite its over-the-counter nature and modest level of trading activity. Specifically, they find that transaction prices are typically quite close to widely available end-of-day quoted prices and that realized bid-ask spreads are modest, even though the reasonably comprehensive data set from 2010 contains just over two trades per day on average. The authors also identify concentrations of activity in certain tenors (ten years) and trade sizes ($25 million) and among certain market participants, as well as various attributes that help explain trade sizes and price deviations. Their study can serve as a resource for policymakers considering public reporting and other regulatory initiatives and for market participants and observers more generally interested in the workings of the inflation swap market.

Although China now has one of the largest government bond markets in the world, the market has received relatively little attention and analysis. We describe the history and structure of the market and assess its functioning. We find that trading in individual bonds was historically sparse but has increased markedly in recent years. We find also that certain announcements of macroeconomic news, such as China’s producer price index (PPI) and manufacturing purchasing managers’ index (PMI), have significant effects on yields, even when such yields are measured at a daily level. Despite the increased activity in the market, we are able to reject the null hypothesis of market efficiency under two different tests for four of the most actively traded bonds.

This paper studies the workup protocol, a unique trading feature in the U.S. Treasury securities market that resembles a mechanism for discovering dark liquidity. We quantify its role in the price formation process in a model of the dynamics of price and segmented order flow induced by the protocol. We find that the dark liquidity pool generally contains less information than its transparent counterpart, but that its role is not trivial. We also show that workups are used more often around volatile times, but that their information role becomes relatively less important at those times compared to that of pre-workup trades. Higher usage of workups is also associated with higher market depth, lower bid-ask spreads, and higher trading intensity. Collectively, the evidence suggests that workups tend to be used more as a channel for liquidity providers to guard against adverse price movements than as a channel to hide private information.

We examine the relationship between monetary policy operations and interbank borrowing and lending of funds using sovereign bonds as collateral. We first establish that, in the precrisis period, there are important but rather weak relations between these funding sources and that this relationship varies within maintenance periods and at the end of the year. Official funding conditions did not meaningfully constrain repo market activity in the 2003-05 period but, in the immediate precrisis period, rate increases led to a sharp contraction in repo activity. Focusing on the crisis period, we identify potentially benign substitution effects between official auctions and repo market activity but our empirical analysis shows that positive innovations in the cost of official funding, due to aggressive bidding, and a limited allotment response, encouraged increased use of the interbank repo market. The analysis informs a discussion of the merits of returning to variable rate operations.

The potential advantages from the introduction of Treasury inflation-protected securities (TIPS) in 1997 have not been fully realized, mainly because TIPS are less liquid than nominal Treasury securities. The lack of liquidity is thought to adversely affect TIPS prices relative to prices of nominal securities, offsetting the benefits that come from TIPS having no inflation risk. Despite the importance of TIPS liquidity and the market’s large size, there is virtually no quantitative evidence on the securities’ liquidity. This article sheds light on this phenomenon using novel tick data from the interdealer market. The authors identify several features of the TIPS market also present in the nominal securities market, but some unique features as well. As in the nominal market, there is a marked difference in trading activity between the most recently issued (“on-the-run”) and previously issued (“off-the-run”) securities, as trading drops sharply when securities go off the run. In contrast to the nominal market, there is little difference in bid-ask spreads or quoted depth between these securities, but there is a difference in the incidence of posted quotes. These results suggest that trading activity and quote incidence may be better cross-sectional measures of liquidity in the TIPS market than bid-ask spreads or quoted depth. Intraday patterns of trading activity are broadly similar in both markets, but TIPS activity peaks somewhat later, likely reflecting differences in the use and ownership of these securities. Announcement effects also differ between markets, with TIPS auction results and CPI releases eliciting particularly strong increases in trading activity, likely indicating these announcements’ special importance to TIPS valuation.

We propose a new class of dynamic order book models that allow us to 1) study episodes of extreme low liquidity and 2) unite liquidity and volatility in one framework through which their joint dynamics can be examined. Liquidity and volatility in the U.S. Treasury securities market are analyzed around the time of economic announcements, throughout the recent financial crisis, and during flight-to-safety episodes. We document that Treasury market depth declines sharply during the crisis, accompanied by increased price volatility, but that trading activity seems unaffected until after the Lehman Brothers bankruptcy. Our models’ key finding is that price volatility and depth at the best bid and ask prices exhibit a negative feedback relationship and that each becomes more persistent during the crisis. Lastly, we characterize the Treasury market during flights to safety as having much lower market depth, along with higher trading volume and greater price uncertainty.

This paper examines the over-the-counter (OTC) interest rate derivatives (IRD) market in order to inform the design of post-trade price reporting. Our analysis uses a novel transaction-level data set to examine trading activity, the composition of market participants, levels of product standardization, and market-making behavior. We find that trading activity in the IRD market is dispersed across a broad array of product types, currency denominations, and maturities, leading to more than 10,500 observed unique product combinations. While a select group of standard instruments trade with relative frequency and may provide timely and pertinent price information for market participants, many other IRD instruments trade infrequently and with diverse contract terms, limiting the impact on price formation from the reporting of those transactions. Nonetheless, we find evidence of dealers hedging rapidly after large interest rate swap trades, suggesting that, for this product, a price-reporting regime could be designed in a manner that does not disrupt market-making activity.

This review examines the Federal Reserve's (or Fed's) unprecedented liquidity provision during the financial crisis of 2007–2009. It first reviews how the Fed provides liquidity in normal times. It then explains how the Fed's new and expanded liquidity facilities were intended to enable the central bank to fulfill its traditional lender-of-last-resort role during the crisis while mitigating stigma, broadening the set of institutions with access to liquidity, and increasing the flexibility with which institutions could tap such liquidity. The review then assesses the growing empirical literature on the effectiveness of the facilities and provides insights as to where further research is warranted.

Ongoing regulatory reform efforts aim to make the over-the-counter derivatives market more transparent by introducing public reporting of transaction-level information, including price and volume of trades. However, to date there has been a scarcity of data on the structure of trading in this market. This paper analyzes three months of global credit default swap (CDS) transactions and presents findings on the market composition, trading dynamics, and level of standardization. We find that trading activity in the CDS market is relatively low, with a majority of reference entities for single-name CDS trading less than once a day. We also find that a high proportion of CDS transactions conform to standardized contractual and trading conventions. Examining the dealer’s role as market maker, we find that large trades with customers are generally not rapidly offset by further trades in the same reference entity, suggesting that hedging of large positions, if taking place, occurs over a longer time horizon. Through our analysis, we provide a framework for regulators and policymakers to consider the design of the public reporting regime and the necessary improvements to data collection to facilitate meaningful price reporting for credit derivatives.

One of the chief actions taken by the Federal Reserve in response to the financial crisis was the introduction or expansion of facilities designed to provide liquidity to the funding markets. A study of the programs suggests that the liquidity facilities generated $20 billion in interest and fee income between August 2007 and December 2009, or $13 billion after taking into account the estimated $7 billion cost of funds. Moreover, the Fed took important steps to limit the credit exposure it incurred in connection with the facilities.

The financial crisis that began in August 2007 disrupted U.S. dollar funding markets not only in the United States but also overseas. To address funding pressures internationally, the Federal Reserve introduced a system of reciprocal currency arrangements, or “swap lines,” with other central banks. The swap line program, which ended early this year, enhanced the ability of these central banks to provide U.S. dollar funding to financial institutions in their jurisdictions.

The financial crisis that began in August 2007 disrupted U.S. dollar funding markets not only in the United States but also overseas. To address funding pressures internationally, the Federal Reserve introduced a system of reciprocal currency arrangements, or "swap lines," with other central banks. The swap line program, which ended early this year, enhanced the ability of these central banks to provide U.S. dollar funding to financial institutions in their jurisdictions.

The Federal Reserve launched the Term Securities Lending Facility (TSLF) in 2008 to promote liquidity in the funding markets and improve the operation of the broader financial markets. The facility increases the ability of dealers to obtain cash in the private market by enabling them to pledge securities temporarily as collateral for Treasuries, which are relatively easy to finance. The TSLF thus reduces the need for dealers to sell assets into illiquid markets as well as lessens the likelihood of a loss of confidence among lenders.

We assess the microstructure of the U.S. Treasury securities market following its migration to electronic trading. We model price discovery using a vector autoregression model of price and order flow. We show that both trades and limit orders affect price dynamics, suggesting that traders also choose limit orders to exploit their information. Moreover, while limit orders have smaller price impact, their greater variation contributes more to the variance of price updates. Lastly, we find increased price impact of trades and especially limit orders following major announcements (such as FOMC rate decisions and macroeconomic data releases), suggesting that the private information derived from public information is disproportionally exploited through limit orders.

Using data on U.S. Treasury dealer positions from 1990 to 2006, we find evidence of a significant role for dealers in the intertemporal intermediation of new Treasury security supply. Dealers regularly take into inventory a large share of Treasury issuance so that dealer positions increase during auction weeks. These inventory increases are only partially offset in adjacent weeks and are not significantly hedged with futures. Dealers seem to be compensated for the risk associated with these inventory changes by means of price appreciation in the subsequent week.

The U.S. Treasury Department now releases fuller information about its auctions than in the past, including new information on investor class and bidder category. The investor class data shed light on the distribution of demand for government securities, and the bidder category data, released first, offer an early read on demand. Purchases by indirect bidders, in particular, are a fairly good proxy for foreign purchases of Treasury notes, but not Treasury bills.

The Federal Reserve now makes available current and historical data on trades in U.S. Treasury and other securities that fail to settle as scheduled. An analysis of the data reveals substantial variation in the frequency of fails over the 1990-2004 period. It also suggests that surges in fails sometimes result from operational disruptions, but often reflect market participants' insufficient incentive to avoid failing.

The Federal Reserve collects data on the financing activities of the primary government securities dealers. Some market analysts argue that the data show a considerable rise in dealer leverage in recent years. However, a close reading of the data suggests that dealer borrowing involving fixed-income securities has grown only modestly. Moreover, the increase that has occurred is not clearly associated with greater risk taking.

Contrary to popular belief, interest rates can drop below zero. From early August to mid-November of 2003, negative rates occurred on certain U.S. Treasury security repurchase agreements. An examination of the market conditions behind this development reveals why market participants are sometimes willing to pay interest on money lent.

We show that Treasury bill auction procedures create classes of price-equivalent discount rates for bills with fewer than seventy-two days to maturity. We argue that it is inefficient for market participants to bid at a discount rate that is not the minimum rate in its class. The inefficiency of bidding at a rate other than the minimum is related to a quantity shortfall rather than an unexploited profit opportunity. Auction results for weekly offerings of four-week bills and occasional offerings of cash management bills show that market participants frequently bid at inefficient rates. However, they are more likely to bid at efficient rates than chance would suggest.

This paper was presented at the conference "Economic Statistics: New Needs for the Twenty-First Century," cosponsored by the Federal Reserve Bank of New York, the Conference on Research in Income and Wealth, and the National Association for Business Economics, July 11, 2002. Securities liquidity is important to those who transact in markets, those who monitor market conditions, and those who analyze market developments. This article estimates and evaluates a comprehensive set of liquidity measures for the U.S. Treasury securities market. The author finds that the commonly used bid-ask spread-the difference between bid and offer prices-is a useful measure for assessing and tracking liquidity. The spread is highly correlated with a more sophisticated price impact measure and is correlated with episodes of reported poor liquidity in the expected manner. He also finds that other measures correlate less strongly with episodes of poor liquidity and with the bid-ask spread and price impact measures, indicating that they are only modest proxies for market liquidity. Trading volume and trading frequency, in particular, are found to be weak proxies for market liquidity, as both high and low levels of trading activity are associated with periods of poor liquidity.

Settlement fails, which occur when securities are not delivered and paid for on the date scheduled by the buyer and seller, can expose market participants to the risk of loss due to counterparty insolvency. This article examines the institutional and economic setting of the fails problem that affected the Treasury market following September 11 and describes how the Federal Reserve and the U.S. Treasury responded. The authors explain that fails rose initially because of the physical destruction of trade records and communication facilities. Fails remained high because a relatively low federal funds rate and investor reluctance to lend securities kept the cost of borrowing securities to avert or remedy a fail comparable to the cost of continuing to fail. The fails problem was ultimately resolved when the Treasury increased the outstanding supply of the on-the-run ten-year note through an unprecedented "snap" reopening. The article also suggests other ways to alleviate chronic fails, such as the introduction of a securities lending facility run by the Treasury and the institution of a penalty fee for fails.

This paper makes use of a natural experiment of the U.S. Treasury Department to examine the relationship between Treasury security issue size and liquidity. Treasury bills that were first issued with fifty-two weeks to maturity and then reopened at twenty-six weeks are shown to be more liquid than comparable maturity bills that were first issued with twenty-six weeks to maturity. The relationship is less pronounced when bills are on-the-run (the most recently auctioned bills of a given maturity) than when they are off-the-run, and persists when controlling for other factors that affect liquidity. The reopened bills are found to have higher yields (lower prices) than comparable maturity bills, showing that the indirect liquidity benefits of reopenings are more than offset by the direct supply costs.

The market for U.S. Treasury securities operates around-the-clock from the three main trading centers of Tokyo, London, and New York. We examine this market for volatility spillovers using the methodology employed by Engle, Ito, and Lin (1990) for the foreign exchange market. We find meteor showers in Tokyo and London but not New York; i.e., volatility spills over into Tokyo and London from the other trading centers, but not into New York. We also find that lagged trading volume significantly impacts U.S. Treasury yield volatility for the overseas trading centers, although it does not change the basic meteor shower findings.

The arrival of public information in the U.S. Treasury market sets off a two-stage adjustment process for prices, trading volume, and bid-ask spreads. In a brief first stage, the release of a major macroeconomic announcement induces a sharp and nearly instantaneous price change with a reduction in trading volume, demonstrating that price reactions to public information do not require trading. The spread widens dramatically at announcement, evidently driven by inventory control concerns. In a prolonged second stage, trading volume surges, price volatility persists, and spreads remain moderately wide as investors trade to reconcile residual differences in their private views. How DO FINANCIAL MARKETS FORM PRICES and provide liquidity when information arrives? Efforts to address this issue have led to an extensive microstructure literature on equity markets, and to a lesser extent on the foreign exchange (FX) market. Relatively little work has focused on the U.S. Treasury market, however, in spite of its importance in its own right and the advantages it offers as an alternative testing ground for microstructure hypotheses.

We analyse high-frequency responses of the US yield curve to macroeconomic announcements, exploiting the high signal-to-noise ratios of these events. Surprises in the announcements evoke relatively weak reactions from the short maturities and the strong ones from the intermediate maturities. Thus the term structure of announcement effects is hump-shaped. We fit an affine-yield model to the yield changes, using the announcement surprises as instruments for the Generalised Method of Moments (GMM). The model estimates imply that the announcements impose larger shocks on an expected future target interest rate than on the current short-term interest rate and that different types of announcements generate different expectations about this target rate, how rapidly it will be approached, and how long it will be maintained.

U.S. Treasury securities are traded in London and Tokyo as well as in New York, creating a virtual round-the-clock market. The author describes that market by examining trading volume, price volatility, and bid-ask spreads over the global trading day. He finds that trading volume and price volatility are highly concentrated in New York trading hours. Bid-ask spreads are found to be wider overseas than in New York and wider in Tokyo than in London. Despite the lower liquidity of the overseas locations, the author finds that overseas price changes in U.S. Treasury securities are unbiased predictors of overnight New York price changes.