## 1. Ownership of U.S. Treasuries

## Source details

**Canonical URL:** [1. Ownership of U.S. Treasuries](https://www.imf.org/-/media/files/publications/wp/2017/wp17172.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2017/wp17172.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2017/wp17172.pdf.json)

---

### I. Introduction — core insights and mechanisms
- Short-term policy rates in many AEs have remained persistently low since the aftermath of the recent financial crisis.
- Several leading central banks (e.g., the Federal Reserve in the U.S., the Bank of England, the European Central Bank, and the Bank of Japan) have carried out several rounds of QE to provide further monetary stimulus.
- QE in AEs has generated documented spillovers to EMs; this paper focuses on a collateral channel of transmission distinct from the traditional financial accelerator.
- Two reasons the collateral channel matters:
  - AE central bank purchases have concentrated on bonds that are widely used as collateral (e.g., UST), which facilitate collateralized cross-border funding (repo, securities lending, prime brokerage, derivatives).
  - QE’s absorption of high-quality collateral from private markets can disrupt collateral market functioning, especially given increased demand for high-quality liquid assets after the crisis.
- Key conceptual result: conventional interest-rate policy and balance sheet adjustments are two independent dimensions of monetary policy in some AEs and have different financial spillovers to EMs.
- QE can:
  - Raise the price of UST via Federal Reserve purchases financed by issuing riskless central bank reserves.
  - Widen international spreads and trigger portfolio shifts by AE agents toward higher-return EMB when QE is sufficiently large, potentially weakening or reversing the marginal price effect on UST.
- Two QE unwind scenarios and EM implications:
  - (i) AE central bank adjusts the short-term policy rate while maintaining its balance sheet: EM central banks can mitigate spillovers by aligning short-term rates with the AE.
  - (ii) AE central bank unwinds its balance sheet by selling UST: there is no simple policy alignment for EMs to offset cross-border financial spillovers and financial stability risks.
- Policy complements for EMs in the unwind-by-selling scenario: capital controls and macro-prudential policy may be viable complements.
- EMs with pegged or quasi-pegged exchange rates (examples cited: Hong Kong and Gulf countries; several countries in the Asian-dollar block) may need to reassess policy tools; the paper does not analyze EMs with flexible exchange rates.
- Implication: with AEs acting along two monetary-policy dimensions, merely aligning short-term rates is insufficient to shield EMs from external monetary spillovers.

### II. Model — structure, assets, and constraints
- Environment:
  - Two countries: AE (illustrated as the U.S.) and EM.
  - Time: discrete, two periods t=0,1; period 1 has two possible states U (Up) and D (Down).
  - Single consumption good C in both economies.
- Assets:
  - Riskless bond B: pays one unit of money in both U and D.
  - Risky long-term government bonds Y_AE and Y_EM with payoffs (d_U_AE, d_D_AE) and (d_U_EM, d_D_EM).
  - Only Y_AE (UST analogue) can serve as collateral to obtain collateralized funding in the cross-border market.
- Collateral premium and scarcity:
  - UST enjoys a collateral premium over EMB due to:
    - (a) capital controls on EM agents limiting UST purchases;
    - (b) increased demand for high quality liquid assets due to QE and new regulations.
- Monetary policy specification:
  - AE central bank implements QE by purchasing Y_AE financed by issuing riskless, interest-bearing central bank reserves; amount purchased denoted y_AE_CB.
  - Interest on reserves determines the riskless return i (conventional interest-rate policy).
  - AE central bank modeled as monetary-fiscal authority collecting taxes in period 1 to retire public debt and reimburse earnings/losses from asset purchases.
  - Fixed exchange rate equal to one between AE and EM implies i = i* by default.
- Private finance and collateralized contracts:
  - All private borrowing must be secured by collateral; claims J specify nominal repayment (j_U, j_D) with j_U = j_D = j and one unit of Y_AE required per claim.
  - Actual delivery in period 1: min{j, p_s d_s_AE}.
- EM policy parameters:
  - Capital control parameter τ enters EM period-0 budget as (1+τ) y_AE_h*.
  - Macro-prudential parameter k* enters collateral requirement y_AE_h* ≥ k* sum_j φ_j_h*.
- Equilibrium defined by allocations and prices clearing goods and financial markets and satisfying constraints; model solved numerically as a system of non-linear equations.

### III. QE effects and QE exit scenarios — asset price dynamics
- QE effects (Figure 2; X-axis = share of Y_AE acquired by AE central bank):
  - Initial AE purchases of UST create excess demand for UST and raise UST price relative to EMB (point A to point B).
  - Widening UST–EMB price gap from A to B reflects tighter global collateral constraints.
  - Large enough QE can lead to widening international spread and portfolio shifts by AE agents toward higher-return EMB (beyond point B), producing a “kink” and a temporary decline in UST price relative to EMB.
- QE exit — policy rate hike (Figure 3a):
  - An increase in the AE short-term interest rate leads to a linear decline in the price of UST.
  - Mechanism: leveraged investors fund purchases with collateralized borrowing; higher short-term rates raise borrowing costs and reduce UST prices linearly.
  - With fixed exchange rate assumption, EM aligning its short-term policy rate with the AE causes EMB price changes to mimic UST price changes (blue line overlaps red line).
- QE exit — balance sheet unwind (Figure 3b):
  - AE sells/releases UST back to the market while keeping short-term policy rate unchanged.
  - Initial effect: price of UST remains robust as AE agents returning from EM absorb increased UST supply; small decline in EMB price accompanies this international portfolio shift.
  - Continued supply of UST can eventually dominate returning investor demand, causing UST price to decline and potentially expedite decline in EMB price.
  - Key difference: interest-rate hike primarily affects UST asset price; balance sheet unwind also alters UST supply and thus cannot be simply countered by EM policy-rate alignment.
- Collateral risk channel:
  - Balance sheet adjustment by the Fed can directly influence UST price despite unchanged short-term policy rate.
  - Because UST is prevalent collateral, increased uncertainty about UST value can raise secured funding costs via higher haircuts or interest rates, transmitting long-end effects to short-term rates.
- Market plumbing:
  - Long-term UST are transformed into short-term instruments via repo, securities lending, rehypothecation, and use as margin for OTC derivatives; these maturity-transformation activities cause long-term treasury yield changes to translate into short-term market rate changes.

### IV. Collateral market plumbing and reuse (Box 2)
- Market mechanics:
  - Securities received as collateral are obtained via reverse-repo, securities borrowing, prime brokerage agreements, and OTC derivative positions.
  - Securities pledged, at mark to market values, may be bonds or equities and are cash-equivalent from a legal perspective (title transfer); they do not have to be AAA/AA rated.
- Size and reuse:
  - Fair value of securities received as collateral that is permitted to be sold or re-pledged by global banks:
    - approximately $10 trillion in 2007
    - about $6 trillion in recent years (end 2016 context)
  - Collateral reuse rate (collateral velocity) estimates:
    - about three as of end 2007
    - about 1.8 as of end 2016
- Implications:
  - Declining reuse rate adversely impacts financial lubrication and can complicate monetary transmission.
  - Unwind of central bank balance sheets (release of good collateral), together with reuse rate changes, can result in short-term market rates diverging (a wedge) from policy rates.
  - Collateral velocity is not under central banks’ control; balance sheet reductions and policy rate hikes are not equivalent in their effects on market rates.
- Operational notes:
  - Triparty Repo (TPR) collateral posted via RRP can only be reused within the Triparty system and cannot be used outside that system for central clearinghouses, bilateral derivatives, bilateral repo market, or delivery against short positions.
  - RRP changes liability composition but does not reduce the Fed’s balance sheet size; selling assets from the Fed’s balance sheet reduces the balance sheet and allows new holders to reuse assets without restriction.
  - Federal Reserve balance sheet rose from roughly $1 trillion (end 2007) to more than $4 trillion by end 2014, owing mainly to about $3.4 trillion of asset purchases; corresponding excess reserves were $2.9 trillion.
  - From October 8, 2008 to December 16, 2015, the Federal Reserve offered banks 25 basis points per annum for deposits (including excess deposits), but paid zero interest on deposits from nonbanks, especially GSEs.
  - Since liftoff, the Federal Reserve is offering 125 basis points to banks (IOER) and 100 basis points to eligible nonbanks via the RRP.
  - The General Collateral Finance (GCF) rate approximates market-driven secured funding rates relevant for cross-border pledged collateral flows of almost $6 trillion.
  - With over $2 trillion of excess reserves presently with the banking system, only a genuine balance sheet unwind will reduce excess reserves and help realign GCF with FF.

### V. Policy options for EMs to mitigate balance-sheet-unwind spillovers
- Two policy tools examined:
  - Capital controls (capital flow management):
    - Increasing the EM’s capital control parameter τ puts downward pressure on the price of UST relative to EMB (i.e., strengthens EMB relative to UST).
  - Macro‑prudential policy (raising collateral requirements k*):
    - Raising collateral requirements discourages leveraged investors from purchasing UST, putting downward pressure on the price of UST relative to EMB (i.e., strengthens EMB relative to UST).
- Dynamic response:
  - Varying the intensity of capital controls or macro‑prudential policy in response to an AE central bank’s balance sheet unwind can mitigate most of the relative price effects of the unwind:
    - Capital control parameter modeled as positively correlated with the wedge created by the AE central bank’s balance sheet unwind; tightened when the price gap widens.
    - Macro‑prudential adjustments can be implemented to counteract relative price effects.
- Key takeaway:
  - EMs that peg to AEs may need to be equipped with capital flow management and macroprudential tools because short-term rate alignment alone is insufficient when AE balance sheets are adjusted.

### VI. Key numeric facts and parameters (as provided)
- Federal Reserve balance sheet: roughly $1 trillion (end 2007) to more than $4 trillion (end 2014).
- Asset purchases on Fed asset side: about $3.4 trillion.
- Excess reserves corresponding entry: $2.9 trillion.
- Interest on deposits offered by Fed from October 8, 2008 to December 16, 2015: 25 basis points per annum for banks; zero for nonbanks (GSEs).
- Since liftoff: IOER = 125 basis points to banks; RRP = 100 basis points to eligible nonbanks.
- Cross-border pledged collateral flows approximate: almost $6 trillion.
- Fair value of securities received as collateral permitted to be sold or re-pledged by global banks:
  - approximately $10 trillion in 2007
  - about $6 trillion in recent years (end 2016 context)
- Collateral reuse rate (velocity):
  - about three as of end 2007
  - about 1.8 as of end 2016
- Example arithmetic cited: consensus expects a 3% target (300 bps on $2.5 trillion = $75 billion); 25 bps on $3 trillion = $7.5 billion.

*Source: wp17172 — IMF Working Paper (selected sections: Glossary, I–IV intro; Boxes 1–2).*

### 1. Ownership of U.S. Treasuries ........................................................................................

### 1. Ownership of U.S. Treasuries ................................................................................................6

### 2. Effects of QE on Asset Prices
- Section present in source.

### 3a. Effects of QE Exit: Interest Rate Hike
- Section present in source.

### 3b. Effects of QE Exit: Balance Sheet Unwind
- Section present in source.

### 4. Policy Rates and Market Short-term Rates
- Section present in source.

### 5. After Fed Liftoff: Spread between Repo Rate and Fed Funds has Increased
- Section present in source.

### 6a. Effects of EM's Capital Control
- Section present in source.

### 6b. Effects of EM's Macro-prudential Policy
- Section present in source.

### 7a. Varying Capital Control in Response to Balance Unwind
- Section present in source.

### 7b. Varying Macro-prudential Policy in Response to Balance Sheet Unwind
- Section present in source.

### Boxes
- Box 1. Fed’s Liftoff and the Triparty Structure — Box present in source.
- Box 2. The Financial Plumbing: Pledged Collateral (and its Reuse) Market — Box present in source.

*Source: wp17172 - 1. Ownership of U.S. Treasuries (PDF chapter/section)*

### References .............................................................................................................

### wp17172 - References

### Glossary
- AE: Advanced Economies
- ASW: Araujo, Schommer, and Woodford (2015)
- CFM: Capital Flow Measures
- DTCC: Depository Trust and Clearing Corporation
- EM: Emerging Markets
- EMB: Emerging Market Bonds
- FF: Federal Funds
- FFR: Federal Funds Rate
- FGP: Fostel, Geanakoplos, and Phelan (2017)
- GFC: General Collateral Finance
- GSD: Government Securities Division
- HK: Hong Kong
- IOER: Interest on excess reserves
- OTC: Over-the-Counter
- QE: Quantitative Easing
- RRP: Reverse repo purchase
- SOMA: System Open Market Account
- U.S.: United States
- UST: United States Treasuries

### I. Introduction — core insights and mechanisms
- Short-term policy rates in many AEs have remained persistently low since the aftermath of the recent financial crisis.
- Several leading central banks (e.g., the Federal Reserve in the U.S., the Bank of England, the European Central Bank, and the Bank of Japan) have carried out several rounds of QE to provide further monetary stimulus.
- QE in AEs has generated documented spillovers to EMs; the paper focuses on a collateral channel of transmission distinct from the traditional financial accelerator.
- Two reasons the collateral channel matters:
  - AE central bank purchases have concentrated on bonds that are widely used as collateral (e.g., UST), which facilitate collateralized cross-border funding (repo, securities lending, prime brokerage, derivatives).
  - QE’s absorption of high-quality collateral from private markets can disrupt collateral market functioning, especially given increased demand for high-quality liquid assets after the crisis.
- Key conceptual result: conventional interest-rate policy and balance sheet adjustments are two independent dimensions of monetary policy in some AEs and have different financial spillovers to EMs.
- QE can:
  - Raise the price of UST via Federal Reserve purchases financed by issuing riskless central bank reserves.
  - Widen international spreads and trigger portfolio shifts by AE agents toward higher-return EMB when QE is sufficiently large, potentially weakening or reversing the marginal price effect on UST.
- Two QE unwind scenarios and EM implications:
  - (i) AE central bank adjusts the short-term policy rate while maintaining its balance sheet: EM central banks can mitigate spillovers by aligning short-term rates with the AE.
  - (ii) AE central bank unwinds its balance sheet by selling UST: there is no simple policy alignment for EMs to offset cross-border financial spillovers and financial stability risks.
- Policy complements for EMs in the unwind-by-selling scenario: capital controls and macro-prudential policy may be viable complements.
- EMs with pegged or quasi-pegged exchange rates (examples cited: Hong Kong and Gulf countries; several countries in the Asian-dollar block) may need to reassess policy tools; the paper does not analyze EMs with flexible exchange rates.
- The paper notes EM central banks have little control over their domestic yield curve (Naudon and Yany, 2016).
- Implication: with AEs acting along two monetary-policy dimensions, merely aligning short-term rates is insufficient to shield EMs from external monetary spillovers.

### II. Literature review — theoretical and empirical context
- Empirical literature documents sizeable cross-border spillovers from AE QE (examples and findings cited):
  - Fratzscher et al. (2011): earlier QE phases have stronger cross-border asset-price effects; capital flowed out of EMs to the U.S. under QE 1, reversed under QE 2.
  - Chen et al. (2015): QE in the U.S. had more pronounced impacts on EMs than on other AEs, with heterogeneity across countries.
  - Cho and Rhee (2013): more open and developed capital markets experienced greater inflow swings during QE episodes; more stable exchange rates tended to see greater asset price inflation.
- Identification of QE spillovers is challenging due to confounders (relative growth prospects, global risk aversion, endogeneity).
- Theoretical literature on collateral and QE:
  - Builds on ASW (2015), FGP (2017), and Geanakoplos and Wang (2017).
  - ASW studies central bank purchases of collateral-like assets and interference with private collateral constraints.
  - FGP shows financial integration arising from international sharing of scarce collateral.
  - Geanakoplos and Wang analyze central bank purchases of collateral-like assets and collateralized cross-border funding transmission.
  - Collateral equilibrium models trace to Geanakoplos (1997) and Geanakoplos and Zame (2013).
- BIS report cited: Central Bank Operating Frameworks and Collateral Market, CGFS Publications No. 53.
- Literature on EM policy tools:
  - Forbes et al. (2015): certain macroprudential policies improve measures of financial fragility; capital controls can reduce private credit growth.
  - Blanchard (2016): in a two-country Mundell–Fleming model, capital control more effective than foreign exchange intervention for achieving macro outcomes.
  - IMF (2014) and IMF (2015) discuss spillovers after the Taper Tantrum and the IMF institutional view on capital flow management and macroprudential measures.
- The paper emphasizes that costs of capital controls and macroprudential tools are country-specific and time-varying; welfare analysis is required to judge implementation.

### III. Model — two-country general equilibrium with collateral constraints and monetary policy
- Structure and timing:
  - Two countries: AE (illustrated as the U.S.) and EM.
  - Time is discrete with two periods: t=0,1.
  - Two possible states in period 1: U (Up) and D (Down).
  - Single consumption good C in both economies.
- Asset types and collateral roles:
  - Two kinds of government bonds in each economy:
    - Riskless bond analogous to short-term government bonds (denoted B); pays one unit of money in both U and D.
    - Risky real long-term government bond analogous to long-term government bonds (denoted Y_AE for AE, Y_EM for EM); payoffs exogenously given by (d_U_AE, d_D_AE) and (d_U_EM, d_D_EM).
  - Only Y_AE (UST analogue) can serve as collateral to obtain collateralized funding in the cross-border market.
  - In equilibrium UST enjoys a collateral premium over EMB; collateral premium sources noted:
    - (a) capital controls on EM agents limiting UST purchases;
    - (b) increased demand for high quality liquid assets due to QE and new regulations.
- Central bank monetary-policy specification:
  - AE central bank can implement QE by purchasing risky Y_AE financed by issuing riskless, interest-bearing central bank reserves; amount purchased denoted y_AE_CB.
  - Interest on reserves determines the riskless return i (conventional interest-rate policy).
  - AE central bank is modeled as a monetary-fiscal authority that collects taxes in period 1 to retire public debt and reimburse earnings/losses from asset purchases.
  - As a finite-horizon model, the AE central bank fixes the price of the consumption good in period 1: {p_s}_{s∈{U,D}}.
  - Fixed exchange rate equal to one between AE and EM implies i = i* by default.
  - EM central bank asset purchases are not considered.
- Private finance and collateralized contracts:
  - All private borrowing must be secured by collateral; financial claims J specify nominal repayment (j_U, j_D) and amount of Y_AE required as collateral; for simplicity j_U = j_D = j and each unit of claim must be secured by one unit of Y_AE.
  - Actual delivery of a financial claim in period 1: min{j, p_s d_s_AE}, where p_s d_s_AE represents collateral value in state s.
  - Agents cannot be coerced into honoring promises except via seizure of collateral.
- Agent problems (AE and EM) — key constraints and policy parameters:
  - AE agent maximization subject to:
    - period-0 budget constraint (equation (1));
    - period-1 state budget constraint (equation (2));
    - collateral constraint y_AE_h ≥ sum_j φ_j_h (equation (3)).
  - EM agent maximization similar, but subject to capital control and macro-prudential restrictions:
    - capital control parameter τ enters period-0 budget as (1+τ) y_AE_h* (equation (4));
    - macro-prudential parameter k* enters collateral requirement y_AE_h* ≥ k* sum_j φ_j_h* (equation (6)).
  - Definitions and aggregates specified: prices p = (p_0,{p_s}), financial claim prices q = {q_j}, asset prices π = (π_AE, π_EM), household riskless holdings μ_h, total public debt μ = sum e_B_h + (1+ i)^{-1} π_AE y_AE_CB, and EM aggregate μ* formula involving τ and π_AE y_AE_h*.
- Equilibrium definition: allocation and prices that clear goods and financial markets and satisfy constraints (conditions (i)–(viii)), including:
  - goods market clearing in period 0 and states U,D (conditions (ii) and (iii));
  - asset holdings and central bank holdings aggregate identities (conditions (iv) and (v));
  - net supply zero for each financial claim j (condition (vi));
  - aggregate money balances and public debt identities (conditions (vii) and (viii)).
- Solution approach: the model is solved numerically as a system of non-linear equations; reference to Geanakoplos and Wang (2017) for implementation details.

### IV. Quantitative results — simulation and evidence (introductory note)
- This section presents the model’s major quantitative results (section begins but specific numerical outcomes and figures are outside the provided content).

*Italic: Source — wp17172 (selected sections: Glossary, I–IV intro), IMF Working Paper content as provided.*

### introduction of QE, the AE central bank is able to directly affect the long-term interest rate

### introduction of QE, the AE central bank is able to directly affect the long-term interest rate

### QE can affect long-term yields even with unchanged short-term rates
- QE by the AE central bank can directly affect the long-term interest rate (i.e., 휋퐴퐸), despite keeping the short-term interest rate unchanged.
- Changes in the price gap between UST and EMB (i.e., 휋퐸푀 − 휋퐴퐸) are analogous to changes in long-term yield gaps and reflect the changing “collateral premium” and market tightness for collateral.
- Important implication: without an EM policy response, changes in long-term yield gaps induced by AE QE would translate into cross-border flows.

### Two dimensions of monetary policy during QE exit and asymmetric spillovers
- AE central bank can deploy two policy dimensions during QE exit:
  - (a) adjustment of the short-term policy rate; and
  - (b) adjustment of balance sheet holdings (e.g., unwind).
- These two dimensions differ in financial spillovers to EM:
  - If AE adjusts its short-term policy rate, EM can counter some spillovers by aligning its short-term policy rate with the AE.
  - If AE adjusts its balance sheet (unwinds), there is no simple short-term policy re-alignment for the EM to mitigate changes in long-term yield gaps.

### Model results: asset price dynamics under QE and QE exit
- Effects of QE (Figure 2, X-axis = share of 푌퐴퐸 acquired by the AE central bank):
  - Initial AE purchases of UST create excess demand for UST and raise UST price relative to EMB (point A to point B).
  - Widening UST–EMB price gap from A to B reflects tighter global collateral constraints.
  - Large enough QE can lead to widening international spread and portfolio shifts by AE agents toward higher-return EMB (beyond point B), producing a “kink” and a temporary decline in UST price relative to EMB.
- QE exit — policy rate hike (Figure 3a):
  - An increase in the AE short-term interest rate leads to a linear decline in the price of UST.
  - Mechanism: leveraged investors fund purchases with collateralized borrowing; higher short-term rates raise borrowing costs and reduce UST prices linearly.
  - With a fixed exchange rate assumption, EM aligning its short-term policy rate with the AE causes EMB price changes to mimic UST price changes (blue line overlaps red line).
- QE exit — balance sheet unwind (Figure 3b):
  - AE sells/releases UST back to the market while keeping short-term policy rate unchanged.
  - Initial effect: price of UST remains robust as AE agents returning from EM absorb increased UST supply; small decline in EMB price accompanies this international portfolio shift.
  - Continued supply of UST can eventually dominate returning investor demand, causing UST price to decline and potentially expedite decline in EMB price.
  - Key difference: interest-rate hike primarily affects UST asset price; balance sheet unwind also alters UST supply and thus cannot be simply countered by EM policy-rate alignment.

### Policy options for EMs to mitigate balance-sheet-unwind spillovers
- EM policy tools discussed to narrow long-term yield gaps and mitigate spillovers:
  - Capital controls (capital flow management).
  - Macro-prudential tools.
- These tools may help reduce changes in long-term yield gaps that arise from AE balance-sheet unwind, since simple short-term rate alignment is insufficient in that scenario.

### Collateral risk channel and market plumbing
- Balance sheet adjustment by the Fed can directly influence UST price despite unchanged short-term policy rate.
- Because UST is a prevalent collateral, increased uncertainty or risks about UST value can percolate to short-term collateralized funding rates:
  - Lenders may demand a higher interest rate or higher haircut to compensate for higher probability of default on collateralized contracts.
  - Thus, Fed influence at the long-end can transmit to the short-end via a collateral risk channel.
- Market plumbing transforms long-term UST into short-term instruments via repo, securities lending, rehypothecation, use as margin for OTC derivatives, etc.; these maturity-transformation activities make long-term treasury yield changes translate into short-term market rate changes.
- Collateral reuse/velocity is exogenous to central banks; to control reuse, central banks may use “reverse repo program (RRP)-type structures” to keep collateral velocity muted.
- Bilateral pledged collateral market rates are unobservable but pass through to other interest rates; when the money/collateral nexus works, the General Collateral Finance (GCF) rate proxies bilateral repo rates and thus secured funding rates.

### Fed liftoff, reserves, and secured funding rates
- Due to QE, the Federal Reserve’s balance sheet increased from roughly $1 trillion (end 2007) to more than $4 trillion by end 2014, owing mainly to about $3.4 trillion of asset purchases on its asset side.
- The approximate corresponding entry was excess reserves of $2.9 trillion on the liabilities side.
- From October 8, 2008 to December 16, 2015, the Federal Reserve offered banks 25 basis points per annum for their deposits (including excess deposits over the required reserves), but paid zero interest on deposits from nonbanks, especially GSEs.
- Since liftoff, the Federal Reserve is offering 125 basis points to banks (interest on excess reserves, IOER) and 100 basis points to eligible nonbanks via the reverse repo program (RRP).
- The wedge between GCF and FF (federal funds) matters for policymakers:
  - Choices: (a) focus on the policy rate (volumes shrinking to well below $100 billion/day), where the rate is supported by the RRP floor and IOER ceiling; or (b) focus on the GCF as an approximation of market-driven secured funding rates relevant for cross-border pledged collateral flows of almost $6 trillion.
- Prior to Lehman, Fed interventions via SOMA repo operations addressed general reserve shortages so that the Fed Funds rate remained aligned with the collateral rate (GCF). Post-QE excess reserves limit the ability of reserve changes to align these rates; Figure 5 shows a shift to a situation where GCF is permanently higher.

### Operational note on Triparty Repo and RRP (Box 1)
- Triparty Repo (TPR) collateral is operationally constrained: collateral posted via RRP can only be reused within the Triparty system and cannot be used outside that system for central clearinghouses, bilateral derivatives, bilateral repo market, or delivered against short positions.
- The RRP’s operational structure implies that use of RRP does not reduce the Fed’s balance sheet size; it changes the composition of liabilities (i.e., “accounting drainage”) rather than allowing assets to move to final holders who can freely reuse them.
- Selling assets from the Fed’s balance sheet reduces the balance sheet and allows new holders to reuse assets without restriction; moving assets between Fed liability line-items (e.g., reserve balances to RRPs) does not have the same effect.

*Source: wp17172 - introduction of QE, the AE central bank is able to directly affect the long-term interest rate*

### Box 2. The Financial Plumbing: The Pledged Collateral (and its Reuse) Market

### Box 2. The Financial Plumbing: The Pledged Collateral (and its Reuse) Market

### Market structure and mechanics
- Financial agents that settle daily margins may post cash or securities depending on which is “cheapest to deliver.”
- Securities received as collateral are obtained via reverse-repo, securities borrowing, prime brokerage agreements, and over-the-counter (OTC) derivative positions.
- Securities pledged, at mark to market values, may be bonds or equities, are cash-equivalent from a legal perspective (i.e., with title transfer) and do not have to be AAA/AA rated.

### Size of the pledged collateral market and reuse (collateral velocity)
- The fair value of securities received as collateral that is permitted to be sold or re-pledged by global banks:
  - approximately $10 trillion in 2007
  - about $6 trillion in recent years (end 2016 context)
- Using the methodology of Singh (2011) and incorporating “source collateral,” the collateral reuse rate (or “collateral velocity”):
  - about three as of end 2007
  - about 1.8 as of end 2016
- The decline in the reuse rate adversely impacts financial lubrication in the market.

### Implications for central banks and monetary transmission
- Central banks should monitor the collateral reuse rate in the bilateral pledged collateral market, along with money metrics, to gauge the short-term rate environment.
- Unwind of central bank balance sheets (i.e., release of good collateral), together with reuse rate changes, can result in short-term market rates diverging (a wedge) from policy rates.
- For effective monetary policy transmission, market rates need to move in tandem with policy rates.
- Collateral velocity (reuse of collateral when released to the market) is not under central banks’ control; when the Federal Reserve unwinds, this could lead to a larger wedge between short-term repo rates and policy rates.

### Relevant operational and historical points
- The Federal Reserve never remunerated excess reserves prior to the Lehman demise but began doing so on October 8, 2008, after changing the law.
- A shortage of reserves in the banking system is required for the General Collateral Financing (GCF) rate to be in sync with the Federal Funds (FF) rate; with over $2 trillion of excess reserves presently with the banking system, only a genuine balance sheet unwind will reduce excess reserves.
- Singh (2017b) argues that excess reserves and good collateral are different, and that dealer balance sheet space (due to an unwind of the Fed’s balance sheet) may increase collateral reuse and have an easing effect; balance sheet reductions and policy rate hikes are not equivalent.
- The arithmetic of remunerating excess reserves example (as presented):
  - consensus expects a 3% target (300 bps on $2.5 trillion, at present, is $75 billion; 25 bps on $3 trillion, a couple of years ago, was $7.5 billion).

### Policy options for emerging markets (EMs)
- EMs that peg to advanced economies (AEs) may need to be equipped with tools in case the AE they peg to decides to unwind its balance sheet.
- Two policy tools examined:
  - Capital controls in the EM:
    - Increasing the EM’s capital control parameter puts downward pressure on the price of UST relative to EMB (i.e., strengthens EMB relative to UST).
  - Macro‑prudential policy in the EM (modeled as raising collateral requirements on agents):
    - Raising collateral requirements discourages leveraged investors from purchasing UST, putting downward pressure on the price of UST relative to EMB (i.e., strengthens EMB relative to UST).
- Both capital control and macro‑prudential policy can be used to narrow relative price gaps between UST and EMB.
- Varying the intensity of capital controls or macro‑prudential policy in response to an AE central bank’s balance sheet unwind can mitigate most of the relative price effects of the unwind:
  - Capital control parameter is modeled as positively correlated with the wedge created by the AE central bank’s balance sheet unwind; tightening when the price gap widens.
  - Macro‑prudential adjustments can be implemented to counteract relative price effects.

### Key takeaways
- Declining collateral reuse and reductions in pledged collateral quantities reduce market lubrication and can complicate monetary transmission.
- EMs should complement their financial stability toolkits with macroprudential and capital flow management measures, recognizing that balance sheet reductions and policy rate hikes are not equivalent.
- Understanding market signals such as repo rates remains crucial because a normal liftoff assumes all short‑term rates move with the policy rate; if they do not, monetary transmission could be compromised.

*Source: Box 2, "The Financial Plumbing: The Pledged Collateral (and its Reuse) Market," wp17172.*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2017/wp17172.pdf_