## wp18258

## Source details

**Canonical URL:** [wp18258](https://www.imf.org/-/media/files/publications/wp/2018/wp18258.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2018/wp18258.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2018/wp18258.pdf.json)

---

### Main mechanism and empirical approach
- Central bank asset purchases affect individual bond scarcity, producing "specialness" that can lower SC repo rates below the deposit facility rate (DFR) (Duffie, 1996).
- Eurosystem PSPP purchases reduce net supply available in the market and make purchased bonds scarcer.
- Scarcity can induce intermediaries to lend cash at a lower rate to obtain specific bonds (to deliver after a short sale, for regulatory reasons, or to enter a repo).
- Data and sample:
  - Two years of repo transactions from Brokertec, covering bonds issued by the seven largest Euro area countries.
  - Daily frequency between 2nd January 2015 and 9th May 2017.
  - Combined Brokertec transactions with proprietary PSPP purchase data.
- Empirical strategy:
  - Panel regressions with bond and country-maturity-time fixed effects to estimate determinants of security-level repo rates and response to PSPP purchases.
  - Instrumental variables (2SLS) exploiting pre-determined PSPP eligibility rules (maturity and timing) to address endogeneity.

### Key empirical findings and quantitative results
- Security-level effect of asset purchases:
  - OLS: "1% of a bond outstanding purchased is associated with a decline of about 0.78 bps in its repo rate."
  - 2SLS (endogeneity-corrected): effect is about six times larger than OLS; instrumented estimate: purchases of 1% of bond outstanding decrease repo rate by about 5 bps (column 4, table 8).
- Aggregate liquidity effects:
  - Excess liquidity contributes to downward pressure on GC repo rates separate from bond-specific scarcity.
  - Quantitative linkage: 1 additional billion euro of excess liquidity lowers the SC repo rate for any given bond by 0.022 bps; implication: 1,000 billion euro of excess liquidity implies an average effect on SC rates of -22 basis points.
- Distributional findings:
  - PSPP has stronger effects at the bottom of the repo-rate distribution. For the 5% of bonds that trade at the lowest repo rates, purchasing 1% of outstanding is associated with a decline of about 2 bps.
- Descriptive examples:
  - If the GC rate is at "-0.40%", a market participant willing to borrow a specific German Bund maturing on "15 Aug. 2027" might have to accept lending cash at "-0.50%".
  - GC Pooling is secured by a basket of "around 3000 securities."
  - Beyond "500 billion euro" of excess liquidity, Italian and Spanish GC remain muted and floored by the DFR; each extra "1 billion euro" of excess liquidity is associated with "-1 bps" in the German and in the French GC.

### Repo market mechanics, GC versus SC, and market segmentation
- Mechanics and definitions:
  - Repo: borrowing and lending cash against collateral; haircut example: borrow "95" in cash against "100" of collateral (haircut "5%").
  - General Collateral (GC): collateral from a pre-specified basket; treated as a money-market rate in literature.
  - Specific Collateral (SC): a specific ISIN is specified; borrowing a specific bond can command a specialness premium (lower remuneration for cash).
- Observed patterns:
  - Multiple GC rates coexist (pan-euro GC pooling and national GC baskets); national GC rates show hierarchy across jurisdictions.
  - Since 2016 GC repo rates have diverged by country: Italy and Spain GC bounded by the DFR; Germany and France GC trade at significantly lower levels.
  - Market segmentation: some GC Pooling participants lack access to the Eurosystem deposit facility (e.g., foreign banks), preventing arbitrage and allowing repo rates below the DFR.
- Relation between GC and SC:
  - GC rates act as the upper envelope of SC rates; when many bonds are valuable in SC, GC rates may not reflect a true risk-free funding rate.
  - Large dispersion of SC rates exists; some bonds sometimes trade "100 basis points" below the country-GC rate.
  - Empirical analysis focuses on SC rates (security-level) rather than GC-SC spread.

### Theoretical mechanism and supply/demand drivers of specialness
- Theoretical channel:
  - Specialness arises when specific securities are actively sought in the repo market (e.g., to cover short sales); specialness premium = GC rate − SC rate.
  - Key friction: some holders lend only at a lower SC rate, generating an upward-sloping supply curve for collateral; demand from short sellers is inelastic for a given size of short positions.
- Eurosystem-specific supply and demand shocks:
  - APP announced 22th January 2015; purchases started March 2015 at around 50 billion euro/month for Euro area government bonds.
  - As of May 2017 the Eurosystem held 1500 billion euro of government bonds, among which around 900 billion euro are rated at least single A.
  - Total Euro area sovereign bond market represents around 7,500 billion euro (ECB).
  - PSPP purchases likely increased short selling by counterparties who sell bonds to the central bank and end up with short positions.
  - Regulatory changes since 2015 (Leverage Ratio, Liquidity Coverage Ratio) may reduce banks' repo intermediation capacity, allowing larger specialness premia to persist.
  - On-the-run and "cheapest-to-deliver" bonds particularly subject to short-selling and specialness.

### Data, sample construction, and summary statistics
- Brokertec dataset:
  - Slightly more than 5 million trades between January 2, 2015 and May 9, 2017; collateralized with 1,282 different ISINs.
  - Focused sample: SC repo transactions for Germany (DE), France (FR), Italie (IT), Spain (ES), Netherlands (NL), Austria (AT), Belgium (BE).
  - SC trades constitute 87% on average of Brokertec transactions during the period.
  - Focus tenor: one-day maturity "spot-next", representing 67% of transactions.
  - Average daily exchanged amount on the period: around 170 bn EUR.
- PSPP variable:
  - For each bond computed cumulative PSPP purchases and ratio of cumulative purchases to nominal outstanding (PSPP variable; if Eurosystem bought 20% of a bond, PSPP = 20).
- Cleaned panel:
  - Cleaned sample: 203,203 observations for 823 individual securities (unbalanced panel).
  - Selected summary statistics (Table 1):
    - Repo rate (bps): mean -40.48; min -153.79; max 10.00; sd 20.42; obs 203203
    - Share held PSPP (%): mean 60; obs 203203
    - Time-to-maturity (yr): mean 6.99; min 0.01; max 70.05; sd 8.59; obs 203203
    - Nom. outstanding (bn Eur): mean 14.52; min 0.00; max 43.19; sd 8.26; obs 203203
    - Share held by inelastic investors: mean 0.21; min 0.00; max 0.90; sd 0.17; obs 199885
    - ISIN-Daily volume in Repo (mn Eur): mean 29; min 1.00; max 7571; sd 4543; obs 203203
    - Credit rating (1=AAA): mean 2.96; min 1.00; max 8.00; sd 2.64; obs 203203

### Empirical strategy and identification
- Main specification:
  - First-difference regressions to address persistence:
    - ∆Repo rate_i,t = β ∆PSPP_i,t + FE_i + FE_country,bucket,t + ε_i,t
  - Heterogeneity: on-the-run and cheapest-to-deliver dummies interacted with ∆PSPP; interaction with share held by inelastic investors and with country dummies.
- Alternative specification:
  - ∆Repo rate_i,t = β1 ∆PSPP_i,t + β2 Excess liquidity_t + β3 OIS_t + β4 VSTOXX_t + β5 CDSs_t + β6 SLF vs cash + β7 End_m,q + FE_i + ε_i,t
  - SLF vs cash dummy captures Securities Lending Facility against cash implementation from Dec 15th 2016 (limited to total usage by counterparties at 50 bn Eur).
- Robustness: standard errors clustered at maturity-issuer level; serial correlation tests and alternative outlier treatments performed.

### Addressing potential reverse causality (instrumental strategy)
- Endogeneity concern:
  - PSPP purchases may have been adjusted to avoid buying securities with special repo rates; OLS likely underestimates true effect.
  - ECB guidance quoted regarding trying to avoid purchasing securities commanding special rates in the repo market.
- Instrument construction (Eligible_i,t):
  - Rule 1: minimum residual maturity (initially two years, then one year from January 2017).
  - Rule 2: maximum residual maturity of 30 years.
  - Rule 3: blackout around issuance/re-issuance dates (primary vs secondary market distinction).
  - Eligible_i,t = 1 when all three rules satisfied; first-stage: ˆPSPP_it = β1 Eligible_it + FE_i + FE_country,bucket,t + ε_it
- 2SLS results and diagnostics:
  - Eligible_i,t is positive and significant in first stage; F statistic = 170.
  - 2SLS estimate: purchases of 1% of a bond outstanding decrease repo rate by about 5 bps (column 4 of table 8), more than sixfold OLS estimate.
  - Stock-Yogo and Kleibergen tests checked to dissipate weak-instrument concerns.
- Robustness:
  - Excluding first semester of 2015 or subsample analyses do not materially change results.

### Policy implications and recommended tools
- Monetary policy transmission and communication:
  - Asset purchases that flatten the yield curve can increase dispersion in short-term money market rates, complicating central bank ability to pass on future policy rate changes across markets.
  - GC rates may deviate from policy-controlled rates and cannot be considered pure risk-free rates when specialness is prevalent.
  - Trade-off: steering short-end (money market rates) versus controlling long-end (yield curve) via asset purchases.
- Operational and regulatory tools:
  - Manage excess liquidity when central bank floor becomes "leaky", including adjusting perimeter of institutions with access to central bank facilities (e.g., opening overnight reverse repo to non-banks).
  - Securities Lending Facility (SLF): lend out purchased bonds to alleviate bond scarcity; an SLF offering bonds at a minimum spread (example: 10 bps) shifts supply and reduces specialness premium.
  - Options include extending access to operational instruments to non-banks, implementing/enlarging SLFs, and harmonizing SLF conventions across national central banks to reduce fragmentation.
- Observed SLF conditions (summarized as presented):
  - Germany: GC rate minus confidential spread; 200 mln EUR/security
  - Netherlands: GC rate minus 10 to 25 basis points; 200 mln EUR/security
  - France: GC rate minus 10 to 25 bps; 200 mln EUR/security
  - Italy: The special repo rate for that security minus an additional penalty of 10 basis points; 200 mln EUR/security
  - Spain: The most penalizing of: special repo market rate for that security or the GC rate minus 10 basis point; 200 mln EUR/security
  - US: Minimum bid rate of 5 basis points; 90% limit on holdings
  - Eurosystem (SLF against cash): -30 basis points below the ECB deposit facility rate; 50bn EUR in total
- Welfare and distributional notes:
  - Specialness imposes costs on agents who need to borrow scarce bonds (e.g., banks meeting regulatory obligations) because they forgo difference between repo rate and DFR.
  - Distributional effects and welfare implications remain unclear and warrant further research.

### Relation to existing literature and broader conclusions
- Consistency with prior work:
  - Connects to Jordan and Jordan (1997), D’Amico et al. (2014), Corradin and Maddaloni (2017), and theoretical contributions from Duffie (1996), Duffie and Krishnamurthy (2016), Bottazzi et al. (2012), Huh and Infante (2017).
- Broader conclusion:
  - PSPP depressed repo rates by intensifying bond scarcity and by increasing excess liquidity.
  - Specialness is a channel through which central bank asset purchases can lower bond yields and affect monetary policy transmission.
  - Central banks possess instruments (SLFs, operational access adjustments) that can mitigate distortions from low repo rates and market segmentation.

*Italic: IMF Working Paper wp18258 — Sections 2.1, 2.3, 3, 4, 4.2, 6.2 (selected content).*

### 2.1 and some anecdotal evidence are presented in order to support its empirical relevance.

### wp18258 - 2.1 and some anecdotal evidence are presented in order to support its empirical relevance.

### Main mechanism and empirical approach
- Paper focuses on the mechanism that central bank asset purchases affect individual bond scarcity, thereby affecting repo rates through "specialness" (Duffie, 1996).
- Eurosystem purchases decrease the net supply available in the market and make purchased bonds, all things equal, scarcer.
- Scarcity can make financial intermediaries willing to lend cash at a lower rate to obtain specific bonds (to deliver after a short sale, for regulatory reasons, or to enter a repo), producing specialness such that repo rates can fall below the deposit facility rate.
- Data and sample:
  - Two years of repo transactions from Brokertec (largest electronic platform for European government bonds repo).
  - Focus on bonds issued by the seven largest Euro area countries.
  - Daily frequency between 2nd January 2015 and 9th May 2017.
  - Combined Brokertec transactions with proprietary data on purchases under the PSPP program.
- Empirical strategy:
  - Panel regressions with bond and country-maturity-time fixed effects to estimate determinants of security-level repo rates and their response to Eurosystem purchases.
  - An instrumental variable approach (2SLS) using pre-determined rules set by the Eurosystem regarding maturity and timing of purchases to address endogeneity (rules are unrelated to the repo market).

### Key empirical findings and quantitative results
- Security-level effect of asset purchases on repo rates:
  - On average, "1% of a bond outstanding purchased is associated with a decline of about 0.78 bps in its repo rate."
- Endogeneity-corrected estimate:
  - A 2SLS regression shows that, once properly estimated, "the effect is found to be 6 times larger than when using OLS."
  - Authors state this is the first estimate of asset purchase effects on specialness free of this bias.
- Aggregate liquidity effects:
  - Excess liquidity also contributes to downward pressure on GC repo rates separate from bond-specific scarcity effects.
- Descriptive observations and examples:
  - Example: If the GC rate is at "-0.40%", a market participant willing to borrow a specific German Bund maturing on "15 Aug. 2027" might have to accept lending cash at "-0.50%".
  - GC Pooling is secured by a basket of "around 3000 securities."
  - Beyond "500 billion euro" of excess liquidity, Italian and Spanish GC remain muted and floored by the DFR; each extra "1 billion euro" of excess liquidity is associated with "-1 bps" in the German and in the French GC.

### Repo market mechanics, GC versus SC, and market segmentation
- Repo market basics:
  - Repo allows borrowing and lending cash against collateral (typically bonds); haircut examples: borrow "95" in cash against "100" of collateral (haircut "5%").
  - General Collateral (GC) repos: collateral delivered from a pre-specified basket; main motivation is prior to lend/borrow cash (seen as a risk-free money market rate in literature).
  - Specific Collateral (SC) repos: a specific security (ISIN) is specified as collateral; borrowing a specific bond can command a specialness premium (lower remuneration for cash).
- Observed hierarchy and dispersion:
  - Multiple GC rates coexist (pan-euro GC pooling and national GC baskets); national GC rates show a clear hierarchy across jurisdictions, reflecting credit risk or services.
  - End-of-quarter and end-of-year periods increase volatility and sometimes produce opposite patterns between core and peripheral countries (suggesting securities swaps at accounting dates).
- Divergence across jurisdictions:
  - Since 2016 GC repo rates have been diverging by country. Italy and Spain GC remain effectively bounded by the deposit facility rate (DFR); Germany and France GC trade at significantly lower levels.
  - Market segmentation: some participants in GC Pooling do not have access to the Eurosystem deposit facility (e.g., foreign banks), preventing arbitrage and allowing repo rates to trade below the DFR.
  - PSPP likely increased cash holdings of non-banks, who were substantial sellers to the program.
- Relationship between GC and SC rates:
  - Large dispersion of SC rates exists; some bonds sometimes trade "100 basis points" below the country-GC rate.
  - GC rates behave as the upper envelope of SC rates; when bonds are widely valuable in SC, GC rates may not reflect a true risk-free funding rate but instead the highest SC rates.
  - Because of this, the paper uses SC rates (security-level) rather than the GC-SC spread for empirical analysis.

### Policy implications and recommended tools
- Monetary policy transmission risk:
  - Asset purchases that flatten the yield curve can increase dispersion in short-term money market rates, making it challenging for the central bank to pass on one-for-one future policy rate changes across markets.
  - Quote from ECB Executive Board member Benoît Cœuré: “there is a risk that, under the current framework, some short-term market rates would not respond fully to changes in our key interest rates or, even if they would, that a continued dispersion of short-term rates would adversely impact the transmission of our monetary policy stance” (Reuters (2018)).
  - Trade-off highlighted: controlling the short-end (money market rates) versus the long-end (yield curve) of the curve.
- Policy instruments to address effects:
  - Management of excess liquidity when the central bank's floor becomes "leaky", including adjusting the perimeter of institutions with access to central bank facilities.
  - Securities lending facility: if individual bond scarcity is the main driver of repo rate deviations, a central bank could lend out the bonds it has purchased to alleviate pressure.
- Channel for asset purchase effects on yields:
  - Existing literature shows lower repo rates imply lower yields for the underlying security (Jordan and Jordan (1997)); thus, asset purchase programs' effects on bond yields can be partly explained by abnormally low repo rates (specialness) triggered by purchases.
  - Specialness is therefore one channel through which central bank asset purchases can lower bond yields.

### Relation to literature
- Bond specialness literature: connects to Jordan and Jordan (1997), Buraschi and Menini (2002), Krishnamurthy (2002); empirical work on central bank purchases and specialness includes D’Amico et al. (2014) for the Fed and Corradin and Maddaloni (2017) for Eurosystem interventions in 2011; results are consistent with this paper.
- Theoretical literature: relies on Duffie (1996) seminal work and contributions from Fisher (2002), Bottazzi et al. (2012), Huh and Infante (2017), Duffie and Krishnamurthy (2016); paper’s results confirm intuitions from this literature.
- Central bank interventions and money market rates literature: relates to post-2008 US federal funds market studies such as Bech and Klee (2011), Martin et al. (2015), Garratt et al. (2015), Armenter and Lester (2017).

*Source: wp18258 - 2.1 and some anecdotal evidence are presented in order to support its empirical relevance.*

### 2.3    The effect of bond scarcity on SC repo rates

### 2.3    The effect of bond scarcity on SC repo rates

### 2.3.1    The theoretical mechanism behind specialness
- Specialness arises when specific securities are actively sought in the repo market (Duffie (1996)), e.g., when a financial intermediary short sells a security and must obtain it to deliver.
- Short sellers can either buy the security in the bond market or borrow the security in the repo market at the SC rate.
- The “specialness premium” is defined as the difference between the GC rate and the SC rate of the security.
- Key friction: some holders are unwilling to lend securities at the GC rate and lend only at a lower SC rate, generating an upward-sloping supply curve for collateral.
- Demand for collateral (reverse repos) from short sellers is inelastic to the SC rate for a given size of short positions.
- Consequence: increases in demand or reductions in supply of a security raise the specialness premium (illustrated in Figure 5 with shift from S to S’ and D to D’ increasing the spec premium from s1 to s2).

Notes and market practices highlighted:
- Failing to return a bond at the repo term can trigger complex processes; in practice repo failures can lead to deferred restitution and sometimes transfer of negative cash bearing costs to the failing counterparty (examples: Euroclear remunerates cash balances over 150 million euro at -70bps since end 2016).
- Link between bond outstanding and quantities available for lending in repo market is nontrivial; “reuse” or “velocity of collateral” affects total supply of collateral.

### 2.3.2    Supply and demand shocks in the SC market
- Eurosystem context:
  - Expanded Asset Purchase Program (APP) announced 22th January 2015 including Public Sector Purchase Program (PSPP).
  - Purchases started March 2015 at a pace of around 50 billion euro a month for Euro area government bonds.
  - As of May 2017 the Eurosystem held 1500 billion euro of government bonds, among which around 900 billion euro are rated at least single A.
  - Total Euro area sovereign bond market represents around 7,500 billion euro (ECB).
  - As of May 2017, the Eurosystem had not purchased any Greek bonds.
- Hypothesized impacts:
  - PSPP purchases likely increased short selling by counterparties who sell bonds to the central bank and end up with short positions.
  - Supply-side reductions in collateral availability may result from regulatory constraints entering into force since 2015 (examples include Leverage Ratio (LR) and Liquidity Coverage Ratio (LCR)) that penalize the use of repos by banks and reduce intermediation capacity.
  - Reduced arbitraging capacity of banks can allow larger specialness premia to persist.
- On special bonds:
  - “On-the-run” bonds and “cheapest-to-deliver” bonds are highlighted as particularly subject to heavy short-selling and specialness.

### 3    Data — Repo transactions and bond characteristics (selected points)
- Repo market context and dataset coverage:
  - Repo market represents around 60% of the money market, compared to less than 5% for the unsecured market (ECB (2017b)).
  - Within the repo market, daily turnover of SC (mostly on MTS and Brokertec) was around 200 billion euro daily in Q1 2017 (Coeuré (2017a)).
  - SC daily turnover is more than 20 times the size of the GC pooling market (Eurex).
  - Brokertec accounts for the largest platform for special trades; SC trades constitute 87% on average of Brokertec transactions during the period and this share increased since January 2015.
  - Focused tenor: one-day maturity “spot-next”, representing 67% of transactions in the dataset.
- Raw data and sample construction:
  - Raw data: slightly more than 5 million trades between January 2, 2015 and May 9, 2017, collateralized with 1,282 different ISINs.
  - Focused sample: SC repo transactions of the seven largest Euro area economies: Germany (DE), France (FR), Italie (IT), Spain (ES), Netherlands (NL), Austria (AT), Belgium (BE).
  - Exclusions: bonds issued by supranationals; securities issued by German agency KFW or French agency CADES are excluded.
  - Most Brokertec transactions are cleared through LCH Repoclear with a public haircut schedule; haircuts are relatively stable during the period and captured via maturity-bucket, issuer country and time fixed effects.
  - For each transaction dataset includes: trade date, term, trade volume, rate, collateral ISIN.
  - Average daily exchanged amount on the period: around 170 bn EUR.
  - For each security a daily repo rate Repo Rate_i,t is computed as the volume-weighted average of transaction-level Repo Rate_i,θ on date t:
    - Repo Rate_i,t = [∑_{θ=1}^n Repo Rate_i,θ ×Volumes_i,θ] / [∑_{θ=1}^n Volumes_i,θ]
  - Distribution of SC repo rates shown by country in Figure 8 (not reproduced here).
- PSPP and holdings data:
  - Proprietary data of all PSPP transactions at the security-day level: trade and settlement dates, book value, nominal amount, ISIN.
  - For each bond computed cumulative PSPP purchases and ratio of cumulative purchases to nominal outstanding (PSPP variable; if Eurosystem bought 20% of a bond, PSPP = 20 for that bond).
  - Securities Holdings Database (SHS) used to compute detention structure among institutional investors; define:
    - Inelastic investors: insurance companies and pension funds, households, non-financial corporations and governments.
    - Elastic investors: banks, mutual funds and the foreign sector.
  - Share of nominal outstanding amounts held by inelastic and elastic groups computed per bond.
- Complementary bond-level variables retrieved from EADB or Bloomberg: nominal outstanding, issue and maturity dates, issuer rating, coupon rate; macro variables from Bloomberg and ECB: daily excess liquidity, OIS rates, VSTOXX, sovereign CDS.

### 3.3    Panel description and summary statistics
- Sample cleaning:
  - Removed first and last percentile of specialness rates to handle outliers.
  - Dropped last week of December 2016 and first week of January 2017 because of heightened volatility at year-end (results virtually unchanged if included).
  - Cleaned sample consists of 203,203 observations for 823 individual securities (unbalanced panel).
- Table 1 — Cleaned sample 2nd Jan 2015 - 9th May 2017 (selected variables; mean, min, max, sd, obs)
  - Repo rate (bps): mean -40.48; min -153.79; max 10.00; sd 20.42; obs 203203
  - Share held PSPP (%): mean 60; obs 203203
  - Time-to-maturity (yr): mean 6.99; min 0.01; max 70.05; sd 8.59; obs 203203
  - Nom. outstanding (bn Eur): mean 14.52; min 0.00; max 43.19; sd 8.26; obs 203203
  - Share held by inelastic investors: mean 0.21; min 0.00; max 0.90; sd 0.17; obs 199885
  - ISIN-Daily volume in Repo (mn Eur): mean 29; min 1.00; max 7571; sd 4543; obs 203203
  - Credit rating (1=AAA): mean 2.96; min 1.00; max 8.00; sd 2.64; obs 203203

### 4    Empirical strategy
- Objective: understand behavior of special repo (SC) rates in the Euro area.
- Left-hand-side variable: SC repo rate for a bond (ISIN) on a given day (constructed from Brokertec).
- Main independent variable: PSPP — ratio of PSPP purchases to total nominal outstanding (expressed in percent; PSPP = 20 if 20% of issue bought).
- Estimation approach:
  - Regressions run in first differences to address persistence (∆ operator).
  - Include bond fixed effects (FE_i) and country-maturity-bucket-time fixed effects (FE_country,bucket,t) to capture bond characteristics, macro variables, haircuts, and issuer information.
  - Standard errors clustered at the maturity-issuer level.
- Baseline OLS specification (first differences):
  - ∆Repo rate_i,t = β ∆PSPP_i,t + FE_i + FE_country,bucket,t + ε_i,t
- Heterogeneity by bond type:
  - Construct on-the-run dummy (1 if security is on-the-run) and off-the-run = 1 − on-the-run; interact with ∆PSPP.
  - Specification:
    - ∆Repo rate_i,t = β1 (on-the-run * ∆PSPP_i,t) + β2 (off-the-run * ∆PSPP_i,t) + FE_i + FE_country,bucket,t + ε_i,t
  - Also construct cheapest-to-deliver (ctd) dummy and interact with ∆PSPP.
  - Interact PSPP with proportion held by inelastic investors and with country dummies to explore heterogeneous effects.
- Alternative specification replacing time fixed effects with macro variables:
  - ∆Repo rate_i,t = β1 ∆PSPP_i,t + β2 Excess liquidity_t + β3 OIS_t + β4 VSTOXX_t + β5 CDSs_t + β6 SLF vs cash + β7 End_m,q + FE_i + ε_i,t
  - SLF vs cash dummy captures Securities Lending Facility against cash implementation from Dec 15th 2016, limited to total usage by counterparties at 50 bn Eur.
  - End_m,q dummy captures accounting reporting dates (end of month and end of quarter) that produce spikes in repo rates.
- Robustness checks:
  - First differencing also reduces concern about serial correlation; tests of serial correlation performed as additional checks.

*Source: wp18258 - 2.3    The effect of bond scarcity on SC repo rates*

### 4.2    Addressing potential reverse causality

### 4.2    Addressing potential reverse causality

### Endogeneity concern and motivation for an instrument
- OLS regressions provide a lower bound of the effect of the PSPP on repo rates but may be biased by endogeneity, in particular reverse causality, which would underestimate the effect of the PSPP on repo rates.
- ECB statements indicate PSPP purchases may take into account specialness. Example quote from Benoit Cœuré: “One key principle underlying the implementation of the PSPP is [...] market neutrality of our operations. [...] To this end, we will take particular care to avoid exacerbating any existing market frictions. More specifically, we will try to avoid, to the extent possible, purchasing specific securities such as current cheapest-to-deliver bonds underlying futures contracts, securities commanding special rates in the repo market as a sign of temporary scarcity, and other assets displaying significant liquidity shortages.”
- To address endogeneity concerns, the paper proposes an instrumental variable exploiting legal and technical PSPP eligibility rules that were set before the program started and thus are plausibly exogenous to repo-market specialness.

### Instrument construction: three eligibility rules
- Rule 1: the Eurosystem may not buy bonds below a certain residual maturity. This minimum maturity was fixed at two years initially (ECB (2015c)) and at one year subsequently, from January 2017 onwards (ECB (2017a)).
- Rule 2: the Eurosystem may not buy bonds that have a residual maturity greater than 30 years (ECB (2015c)).
- Rule 3: the Eurosystem may not buy bonds around their issuance or re-issuance dates. The ECB Governing Council applies a “blackout period” and also applies it to neighboring securities along the yield curve to preserve the distinction between primary and secondary markets (ECB (2015a)).
- The instrument Eligible_i,t equals one whenever the three eligibility rules are satisfied, and 0 otherwise.
- The first-stage specification: ˆPSPP_it = β1 Eligible_it + FE_i + FE_country,bucket,t + ε_it

### Baseline OLS findings (summary)
- OLS coefficients on PSPP purchases (as a share of outstanding amount) are negative and strongly significant across fixed-effect specifications.
- Example magnitude: buying 1% of the total outstanding (that would represent on average 150 million) is associated with a decrease of 0.78 basis points of repo rates (third column of table 4).
- Heterogeneous effects:
  - On-the-run vs off-the-run: on-the-run securities are about twice as affected as off-the-run securities (second column of table 5), consistent with theory and D’Amico et al. (2014).
  - Cheapest-to-deliver securities are more affected than other securities (third column of table 5).
  - Investor-type: purchasing 1% of outstanding for a bond wholly held by inelastic investors leads to a decline of 1.8 basis point; purchasing 1% of a bond wholly held by elastic investors leads to a decline of 0.48 basis points (fourth column of table 5).
  - Country heterogeneity: most affected countries in order are Spain, Italy and Germany; in Spain and Italy purchasing 1% of outstanding could decrease repo rates by greater than 1 basis point (last column of table 5).

### Time-varying factors and additional determinants
- Excess liquidity: 1 additional billion euro of excess liquidity lowers the SC repo rate for any given bond by 0.022 bps. Implication: 1,000 billion euro of excess liquidity would imply an average effect on SC rates of -22 basis points.
- MRO and LTRO: excess liquidity created through MRO and LTRO also depresses repo rates via banks withdrawing collateral from the SC repo market.
- Calendar effects: end-of-month and end-of-quarter dummies have negative and highly significant coefficients, consistent with window-dressing around reporting dates and regulatory considerations (LCR, leverage ratio). These effects can reduce supply on the SC repo market.
- Seasonality controls: month and day variables (month number 1-12 and day number 1-31) and last day of the ECB maintenance period are used to control seasonality.
- SLF vs cash dummy (facility against cash collateral from 15th dec 2016) is associated with alleviated pressure on the repo market, though interpretation should be cautious as other changes may coincide.

### Distributional effects across repo-rate percentiles
- Quantile regressions (p5, p25, p25-50, p50-75, p75 and 95p) show PSPP has a stronger effect at the bottom of the distribution (bonds with largest specialness premia).
- For the 5% of bonds that trade at the lowest repo rates, purchasing 1% of outstanding is associated with a decline of about 2 bps.
- PSPP is less or not statistically significant for bonds at the top of the distribution (lowest specialness premia).

### 2SLS (instrumented) results and diagnostics
- Using Eligible_i,t as instrument for ∆PSPP_i,t in a 2SLS yields substantially larger effects compared to OLS.
- Column 4 of table 8: purchases of 1% of the amount of bond outstanding decrease repo rate by about 5 bps (versus 0.78 bps estimated by OLS) — more than sixfold larger once endogeneity is accounted for.
- First-stage diagnostics:
  - The Eligible_i,t coefficient in the first stage is positive and significant (first column of table 8).
  - F statistic = 170 (well above the rule-of-thumb threshold 10).
  - Stock-Yogo critical values and Kleibergen test checked and dissipate weak-instrument concerns.

### Robustness checks
- Excluding the first semester of 2015 (program starts in March 2015 and rates were volatile) does not materially change results.
- Subsample analyses (e.g., late 2016–early 2017) and alternative outlier treatments leave results little affected.

### Key implications for monetary policy
- PSPP depressed repo rates by intensifying bond scarcity and increasing excess liquidity; both raise monetary policy issues.
- Communication and passthrough:
  - GC rates may deviate from policy-controlled rates and cannot be considered pure risk-free rates when specialness is prevalent, complicating monetary policy communication and pass-through.
  - An increase of the DFR may not be transmitted one-for-one to repo rates; deviation between market rates and policy rate signals a trade-off between steering short rates and controlling long-term yields via asset purchases.
- Effects on bond yields and transmission:
  - Repo rates affect bond yields; central bank purchases can affect bond yields through the repo market, potentially reducing monetary policy passthrough across the yield curve as scarcity prevails.
- Distributional and welfare considerations:
  - Specialness imposes costs on those who need to borrow scarce bonds (e.g., banks borrowing to meet regulatory obligations), as they forgo the difference between the repo rate and the deposit facility rate.
  - The distributional effects and welfare implications remain unclear and warrant further research.

*Source: IMF Working Paper (section 4.2).*

### 6.2    What can central banks do about the low level of repo rates?

### 6.2    What can central banks do about the low level of repo rates?

### Instruments to address repo rates below the deposit facility rate (DFR)
- A purely liquidity-driven rate should not trade below the DFR because market participants should prefer to deposit cash with the central bank.
- When some market participants lack access to the central bank, imperfect competition and lack of arbitrage may result in rates below the DFR.
- Central banks may adjust their operational framework to include instruments accessible to non-banks (example cited: US overnight reverse repo facility opened to mutual funds).
- Some country-specific GC rates may not be completely liquidity-driven when all bonds from a given country trade on special; GC rates may reflect the lowest SC rates of specific bonds that are relatively less in demand in the SC repo market.
- Frictions associated with bond scarcity could be alleviated by central banks to improve monetary policy transmission.

### Securities lending facility (SLF) as a remedy for bond scarcity and specialness
- Theoretical literature supports central banks lending bonds acquired through a securities lending facility (SLF).
- Introduction of an SLF that offers bonds for a minimum spread against GC (example in text: 10 bps) shifts up supply in the hypothetical repo market (private repo + SLF), reducing the equilibrium specialness premium (illustrated decrease from s2 to s3 in Figure 7).
- In practice, some euro area SLFs lend specific securities against GC at a “policy spread” at least equal to 10 bps below the GC rate, with per-security borrowing limits (example: 200 mln EUR/security for several national central banks).

### Empirical observations on SLF and pricing
- The period of implementation of the SLF against cash was associated with lower pressure on the special repo market despite limited size.
- More attractive pricing of SLF against cash facilities likely contributed to this effect.
- Borrowing bonds against cash can be more advantageous than borrowing a scarce bond using another scarce bond where all bonds have value in the SC repo market.
- Euro area national central banks manage securities lending on a decentralized basis with different conventions and penalties across national central banks.
- The US overnight securities lending facility offers specific securities against any general collateral at a fee as low as 4.9 basis points.
- Some Eurosystem central banks lend specific securities on more expensive terms (see Table 2 examples showing GC rate minus 10 to 25 basis points or confidential spread, and 200 mln EUR/security limits).

### SLF conditions summarized (as presented)
- Germany: GC rate minus some fixed (confidential) spread; 200 mln EUR/security
- Netherlands: GC rate minus 10 to 25 basis points; 200 mln EUR/security
- France: GC rate minus 10 to 25 bps; 200 mln EUR/security
- Italy: The special repo rate for that security minus an additional penalty of 10 basis points; 200 mln EUR/security
- Spain: The most penalizing of these two rates: special repo market rate for that security or the GC rate minus 10 basis point; 200 mln EUR/security
- US: Minimum bid rate of 5 basis points; 90% limit on holdings
- Eurosystem (SLF against cash): -30 basis points below the ECB deposit facility rate; 50bn EUR in total (Table 3)

### Policy implications and central bank options
- Central banks have instruments to offset problems associated with low repo rates, including:
  - Extending access to operational instruments to non-banks (e.g., overnight reverse repo facilities).
  - Implementing or enlarging securities lending facilities (SLFs) to lend specific securities acquired through asset purchase programs.
- A larger and more attractively priced SLF can reduce specialness premia because specialness persists largely because some agents are unwilling to lend securities they hold.
- Harmonization or coordination of SLF conventions across national central banks could reduce fragmentation and large cross-country differences in terms and penalties.

### Empirical links to PSPP and broader conclusions (from closing section)
- The paper finds PSPP contributions to depressing repo rates both by increasing scarcity of purchased bonds and by increasing excess liquidity.
- Increased dispersion of repo rates below central bank remuneration of reserves may challenge control over short-term interbank rates and the transmission of monetary policy to bond yields and asset prices.
- Central banks have instruments that could mitigate these distortions.
- The study suggests central bank purchases may be accommodated by short sales and covered in the repo market, improving understanding of mechanisms behind asset purchase programs.

*Source: wp18258 - 6.2 What can central banks do about the low level of repo rates?*

### References

### References

### Bibliographic sources (select)
- Adrian, T., Begalle, B., Copeland, A., and Martin, A. (2013). Repo and Securities Lending. In Risk Topography: Systemic Risk and Macro Modeling, NBER Chapters, pages 131–148. National Bureau of Economic Research, Inc.
- Afonso, G. and Lagos, R. (2015). Trade Dynamics in the Market for Federal Funds. Econometrica, 83:263–313.
- Armenter, R. and Lester, B. (2017). Excess Reserves and Monetary Policy Implementation. Review of Economic Dynamics, 23:212–235.
- Arrata, W. and Nguyen, B. (2017). Price impact of bond supply shocks: Evidence from the eurosystem’s asset purchase program. Banque de France Working Paper Series no. 623.
- Baklanova, V., Caglio, C., Cipriani, M., and Copeland, A. (2016). The use of collateral in bilateral repurchase and securities lending agreements. Staff Reports 758, Federal Reserve Bank of New York.
- Bech, M. L. and Klee, E. (2011). The mechanics of a graceful exit: Interest on reserves and segmentation in the federal funds market. Journal of Monetary Economics, 58(5):415–431.
- Bech, M. L. and Malkhozov, A. (2016). How have central banks implemented negative policy rates? BIS Quarterly Review.
- Berentsen, A., Kraenzlin, S., and Müller, B. (2016). Exit strategies for monetary policy. ECON - Working Papers 241, Department of Economics - University of Zurich.
- Boissel, C., Derrien, F., Örs, E., and Thesmar, D. (2014). Sovereign crises and bank financing: Evidence from the european repo market. Manuscript, HEC Paris.
- Bottazzi, J.-M., Luque, J., and Páscoa, M. R. (2012). Securities market theory: Possession, repo and rehypothecation. Journal of Economic Theory, 147(2):477–500.
- Bucalossi, A. and Scalia, A. (2016). Leverage ratio, central bank operations and repo market. Questioni di Economia e Finanza (Occasional Papers) 347, Bank of Italy, Economic Research and International Relations Area.
- Buraschi, A. and Menini, D. (2002). Liquidity risk and specialness. Journal of Financial Economics, 64(2):243–284.
- Carlson, M., Duygan-Bump, B., Natalucci, F., Nelson, B., Ochoa, M., Stein, J., and den Heuvel, S. V. (2016). The Demand for Short-Term, Safe Assets and Financial Stability: Some Evidence and Implications for Central Bank Policies. International Journal of Central Banking, 12(4):307–333.
- Carpenter, S., Demiralp, S., Ihrig, J., and Klee, E. (2015). Analyzing federal reserve asset purchases: From whom does the fed buy? Journal of Banking & Finance, 52:230–244.
- CGFS (2017). Repo market functioning. Technical report, Bank of International settlements.
- Christensen, J. H. and Krogstrup, S. (2014). Transmission of Quantitative Easing: The Role of Central Bank Reserves. Working Paper Series 2014-18, Federal Reserve Bank of San Francisco.
- Christensen, J. H. E., Lopez, J. A., and Shultz, P. (2017). Is There an On-the-Run Premium in TIPS? Working Paper Series 2017-10, Federal Reserve Bank of San Francisco.
- Coeuré, B. (2017a). Asset purchases, financial regulation and repo market activity. Speech at the ERCC General Meeting, Brussels, 14 November, 2017.
- Coeuré, B. (2017b). Bond scarcity and the ecb’s asset purchase programme. Speech at the Club de Gestion Financiere d’Associes en Finance, Paris, 3 April 2017.
- Corradin, S. and Maddaloni, A. (2017). The importance of being special: repo markets during the crisis. Working Paper Series 2065, European Central Bank.
- D’Amico, S., Fan, R., and Kitsul, Y. (2014). The scarcity value of Treasury collateral: Repo market effects of security-specific supply and demand factors. Finance and Economics Discussion Series 2014-60, Board of Governors of the Federal Reserve System (U.S.).
- De Santis, R. A. and Holm-Hadulla, F. (2017). Flow effects of central bank asset purchases on euro area sovereign bond yields: evidence from a natural experiment. Working Paper Series 2052, European Central Bank.
- Duffie, D. (1996). Special repo rates. The Journal of Finance, 51(2):493–526.
- Duffie, D. and Krishnamurthy, A. (2016). Passthrough efficiency in the fed’s new monetary policy setting. In Designing Resilient Monetary Policy Frameworks for the Future. Federal Reserve Bank of Kansas City, Jackson Hole Symposium.
- Dunne, P. G., Fleming, M. J., and Zholos, A. (2013). ECB monetary operations and the interbank repo market. Staff Reports 654, Federal Reserve Bank of New York.
- ECB (2015a). Account of the monetary policy meeting governing council of the european central bank held in frankfurt am main on wednesday and thursday, 21-22 january 2015.
- ECB (2015b). Decision (eu) 2015/1574 of the european central bank. Official Journal of the European Union.
- ECB (2015c). Decision (eu) 2015/774 of the european central bank of 4 march 2015 on a secondary markets public sector asset purchase programme (ecb/2015/10). Official Journal of the European Union.
- ECB (2015d). Money market survey 2015.
- ECB (2017a). Decision (eu) 2015/774 of the european central bank of 11 january 2017 amending decision (eu) 2015/774 on a secondary markets public sector asset purchase programme (ecb/2017/1). Technical report.
- ECB (2017b). Recent developments in euro area repo markets, regulatory reforms and their impact on repo market functioning. Financial Stability Review November 2017 – Special features.
- Ennis, H. M. and Keister, T. (2008). Understanding monetary policy implementation. Economic Quarterly, (Sum):235–263.
- Fache Rousová, L. and Rodríguez Caloca, A. (2015). The use of securities holdings statistics (shs) for designing new euro area financial integration indicators. Technical report, Bank for International Settlements.
- Faubert, V. and Sode, A. (2013). Are safe assets to become scarcer? Tresor Eco no 177.
- Fernandez, R. (2017). European repo markets: Operating within a new normal. Technical report, Nomura European Rates Insights, 5 April 2017.
- Ferrari, M., Guagliano, C., and Mazzacurati, J. (2016). Collateral scarcity premia in eu repo markets.
- Fisher, M. (2002). Special repo rates: An introduction. Economic Review-Federal Reserve Bank of Atlanta, 87(2):27.
- Fleming, M. J. and Garbade, K. D. (2004). Repurchase agreements with negative interest rates. Current Issues in Economics and Finance, 10(Apr).
- Fleming, M. J., Hrung, W. B., and Keane, F. M. (2010). Repo Market Effects of the Term Securities Lending Facility. American Economic Review, 100(2):591–596.
- Flodén, M. (2018). The riksbank’s balance sheet: How large should it be in the future? Speech at the Swedish House of Finance, 13 April, 2018.
- Fontaine, J.-S. and Garcia, R. (2012). Bond Liquidity Premia. Review of Financial Studies, 25(4):1207–1254.
- Frost, J., Logan, L., Martin, A., McCabe, P. E., Natalucci, F. M., and Remache, J. (2015). Overnight RRP operations as a monetary policy tool: some design considerations. Staff Reports 712, Federal Reserve Bank of New York.
- Fuhrer, L. M., Müller, B., and Steiner, L. (2017). The liquidity coverage ratio and security prices. Journal of Banking & Finance, 75:292–311.
- Garratt, R., Martin, A., McAndrews, J. J., and Nosal, E. (2015). Segregated balance accounts. Staff Reports 730, Federal Reserve Bank of New York.
- Greenwood, R., Hanson, S. G., and Stein, J. C. (2015). A Comparative-Advantage Approach to Government Debt Maturity. Journal of Finance, 70(4):1683–1722.
- Greenwood, R., Hanson, S. G., and Stein, J. C. (2016). The federal reserve’s balance sheet as a financial-stability tool. In Designing Resilient Monetary Policy Frameworks for the Future, Jackson Hole Symposium: Federal Reserve Bank of Kansas City.
- Huh, Y. and Infante, S. (2017). Bond Market Intermediation and the Role of Repo. Finance and Economics Discussion Series 2017-003, Board of Governors of the Federal Reserve System (US).
- Jordan, B. D. and Jordan, S. D. (1997). Special repo rates: An empirical analysis. The Journal of Finance, 52(5):2051–2072.
- Kandrac, J. and Schlusche, B. (2017). Quantitative Easing and Bank Risk Taking: Evidence from Lending. Finance and Economics Discussion Series 2017-125, Board of Governors of the Federal Reserve System (US).
- Klee, E. C., Senyuz, Z., and Yoldas, E. (2016). Effects of Changing Monetary and Regulatory Policy on Overnight Money Markets. Finance and Economics Discussion Series 2016-084, Board of Governors of the Federal Reserve System (U.S.).
- Koijen, R. S., Koulischer, F., Nguyen, B., Yogo, M., et al. (2017). Euro-area quantitative easing and portfolio rebalancing. American Economic Review, 107(5):621–627.
- Kreicher, L. L., McCauley, R. N., and McGuire, P. (2013). The 2011 FDIC assessment on banks managed liabilities: interest rate and balance-sheet responses. BIS Working Papers 413, Bank for International Settlements.
- Krishnamurthy, A. (2002). The bond/old-bond spread. Journal of Financial Economics, 66(2-3):463–506.
- Levels, A. and Capel, J. (2012). Is collateral becoming scarce? evidence for the euro area. Journal of Financial Market Infrastructures, 1(1):29–53.
- Mancini, L., Ranaldo, A., and Wrampelmeyer, J. (2016). The euro interbank repo market. Review of Financial Studies, 29(7):1747–1779.
- Martin, A., McAndrews, J., Palida, A., and Skeie, D. (2015). Federal Reserve Tools for Managing Rates and Reserves. IMES Discussion Paper Series 15-E-08, Institute for Monetary and Economic Studies, Bank of Japan.
- Mersch, Y. (2017). Ructions in the repo market – monetary easing or regulatory squeezing? Speech at the GFF summit, Luxembourg, 26 January 2017.
- Munyan, B. (2015). Regulatory Arbitrage in the Repo Market. Working Papers 15-22, Office of Financial Research, US Department of the Treasury.
- Pelizzon, L., Subrahmanyam, M. G., Tomio, D., and Uno, J. (2018). Central bank-driven mispricing. SAFE Working Paper Series 226, Research Center SAFE - Sustainable Architecture for Finance in Europe, Goethe University Frankfurt.
- Poole, W. (1968). Commercial Bank Reserve Management In A Stochastic Model: Implications For Monetary Policy. Journal of Finance, 23(5):769–791.
- Reuters (2018). ECB says the excess cash it created could backfire: Coeure.
- Singh, M. (2011). Velocity of Pledged Collateral; Analysis and Implications. IMF Working Papers 11/256, International Monetary Fund.
- Spies, M. and Sian, H. (2017). Pspp or regulations: What’s driving repo markets? Technical report, Citi, Euro Rates Strategy Focus, 23 March 2017.
- Stein, J. C. (2012). Monetary Policy as Financial Stability Regulation. The Quarterly Journal of Economics, 127(1):57–95.
- Vari, M. (forthcoming). Monetary policy transmission with interbank market fragmentation. Journal of Money, Credit and Banking.

### A. Regulation and the repo market — key analysis and findings
- Overview:
  - Regulations can affect repo markets by incentivizing holding of “safe assets”, limiting collateral reuse, and restricting or penalizing repo leverage.
  - The appendix details specific regulations and their likely impacts on repo market functioning.

- A.1 Regulatory demand of high-quality liquid assets (HQLAs):
  - The Liquidity Coverage Ratio (LCR), part of Basel III, forces banks to hold High Quality Liquid Assets (HQLAs), mostly government bonds.
  - LCR phasing-in started in 2015 and will become fully effective in 2019.
  - As of December 2016:
    - The stock of HQLA assets held by European banks equals 139.5% of LCR requirements.
    - About 2.5 trn EUR of assets are classified as HQLAs.
    - As of December 2016, banks are required to hold around 1.8 trn EUR of HQLAs under the LCR.
  - Collateral borrowed in a reverse repo transaction is counted in the stock of HQLA assets of a bank if it is eligible as such.
  - Borrowing a HQLA asset via reverse repo does not change the LCR ratio; lending HQLA in repo has no effect on a bank’s LCR.
  - The marginal effect of LCR implementation is expected not to be material on repo rates in the US.
  - Fuhrer et al. (2017) find that for the Swiss securities market, qualifying as a HQLA asset decreases on average the yield of a security by around 4 bps.
  - EMIR (margins for OTC derivatives) entered into force in 2014; margins collected by CCPs are subject to haircuts which increase with risk, limiting practical re-use of IMs and VMs and placing high-quality assets on CCP balance sheets.

- A.2 A limit on the re-use of repo collateral:
  - Re-use increases supply of collateral when borrowed securities are lent again.
  - The UCITS V regulation (enforced March 2016) restricts mutual funds’ re-use: re-use permitted only if a collateral asset of at least the same quality is borrowed in exchange.
  - This restriction is likely to decrease supply of collateral on the repo market.

- A.3 Restrictions or costs on repo leverage:
  - Money Market Funds regulation (ratified April 2017) imposes daily and weekly liquidity criteria and restricts use of repurchase agreements: MMFs can only invest in repos up to a limit of 10% of their assets.
  - The Leverage Ratio (LR) imposes capital requirements based on balance sheet size. Under the LR:
    - Banks must have a minimum leverage ratio of 3% (the minimum starts to kick in only in 2018; banks have published leverage ratio since 2015, making 3% a market reference).
  - CGFS (2017) documents that repos expand bank balance sheets and attract a capital charge under the LR, leading banks to adjust prices or limit supply.
  - Expected effects:
    - Banks likely to scale down repos and repo intermediation activities to comply with LR.
    - Repos can be netted, which removes them from the scope of the LR.
  - Empirical evidence:
    - Baklanova et al. (2016) document reductions in repo activity for dealers in US and UK; reduction stronger for safer collateral than riskier collateral.
    - CGFS (2017) notes that activities with low risk weights are more likely to be affected by balance sheet constraints than by risk-weighted capital requirements.

- A.4 Window-dressing around reporting dates:
  - Regulatory compliance is binding at specific dates (month-ends, quarter-ends, year-ends), creating incentives for window-dressing.
  - Examples:
    - LCR reported at month-ends; LR reporting at quarter-ends.
    - Spies and Sian (2017) observe changes in volumes at quarter-ends and clear seasonality at end-quarter as banks’ repo activity as counterparty fell dramatically.
  - Effects on repo markets:
    - Heightened volatility in repo markets during the last days before reporting dates.
    - In jurisdictions with quarter-end reporting, banks contract repo exposure on these dates, producing short-lived but sharp spikes in repo volumes and prices (CGFS (2017)).
    - Year-end reporting adds additional balance sheet constraints, e.g., contribution to the Single Resolution Fund (SRF) calculated annually based on size of banks’ deposits.

_Italic: IMF Working Paper wp18258 — References (appendix A: Regulation and the repo market)_

---


_Source: https://www.imf.org/-/media/files/publications/wp/2018/wp18258.pdf_