FRFARRF: the unwind begins
12 November 2013
The Fixed-Rate Full-Allocation Reverse Repo Facility has been clarified, but an unanswered question remains, says Oscar Huettner
Image: Shutterstock
On 21 August the Federal Reserve released the minutes of its July meeting. Included in these minutes was the announcement of the creation of a Fixed-Rate Full-Allocation Reverse Repo Facility (FRFARRF). This announcement appeared to catch a large portion of the secured financing markets off-guard and has elicited a number of articles speculating on the effects, intended or unintended, of this new Federal Reserve tool.
Since that initial announcement, the Federal Reserve has posted a significant amount of information on its website, which, when read through, debunks a number of the more cynical interpretations of this new tool, but also leaves a bigger question unanswered.
The FRFARRF allows market participants to ‘reverse in’ collateral from the Federal Reserve, therefore investing cash and, from the Federal Reserve’s point of view, draining liquidity from the banking system. The Federal Reserve created a pilot programme to run from 23 September until 29 January 2014. By its terms, participants can bid for up to $500 million of collateral on an overnight basis (this could be increased to as high as $1 billion). Rates were set at 0.01 and could be raised to as high as 0.05. The window for these operations is between 11:15 and 11:45am EST. Only the Federal Reserve’s treasury securities are to be used.
Currently, 139 institutions are eligible to participate: at present, this includes 21 primary dealers, 18 banks, six government-sponsored enterprises (including Fannie Mae and Freddie Mac) and 94 large 2a-7 funds. This facility contrasts with the Federal Reserve’s traditional reliance on its primary dealer network exclusively to affect short-term monetary policy. The first operation was conducted on 23 September, and subsequently the Federal Reserve announced on 26 September that it would increase the maximum allocation limit to the $1 billion level.
The market can often overreact when a surprise announcement comes from the Federal Reserve. It was immediately pointed out that the central bank would now be conducting open-market operations directly with market participants at the expense of its primary dealer network. Some speculated that the creation of this new facility could partially reflect the Federal Reserve’s dissatisfaction with the pace of triparty reform. Finally, there was the prediction that the central bank’s presence in the overnight markets could raise dealers’ financing costs and serve as another lever to force them to reduce their balance sheets.
While the FRFARRF will undoubtedly alter the short-term funding markets, it is worth reading the Federal Reserve’s comments on the purpose of this facility as it is explained on its website.
The Federal Reserve makes two key points. First, the facility will complement the its payment of interest on overnight excess reserves, allowing the central bank to directly influence short-term secured and unsecured rates. Second, the facility is intended to work in conjunction with the Federal Reserve’s more traditional capped allotment repos. The central bank goes to great lengths to refer to this facility as an operational exercise, but it seems likely that an expanded FRFARRP is intended to play a major role in the Federal Reserve’s transition from its current accommodative stance to one where it will be withdrawing liquidity.
The most obvious consequence of an expanded, permanent FRFARRF would be the creation of a floor under overnight repo rates. If an institutional investor can access collateral from the Federal Reserve at 0.25 in the late morning, the investor will demand a higher rate from its market counterparties in the early am. The interesting question is how much of a premium dealers will have to pay.
While some market commentators have speculated that the Federal Reserve has abandoned its traditional reliance on its primary dealer network, an alternative explanation may be that it recognises that with all of the regulations aimed at limiting dealers’ leverage, they may not be able to play their traditional role of serving as a conduit for the Federal Reserve’s open market operations. Other considerations may be the central bank’s ongoing efforts to ensure the viability of 2a-7 funds and to compensate for collateral shortages around reporting dates.
The big winners appear be Fannie Mae and Freddie Mac, as they can now buy repos from the Federal Reserve, whereas they currently cannot access the IEOR programme. This will further reduce the volume in the overnight Federal Reserve funds market, as Fannie Mae and Freddie Mac currently have no alternative but to place their excess liquidity there.
Now, for the unanswered question: will the Federal Reserve rely exclusively on the overnight repo market to unwind quantitative easing when the time comes to withdraw liquidity?
As of the 25 September, the Federal Reserve held $3.45 trillion of securities in its System Open Market Account (SOMA) programme, $1.96 trillion of treasuries and $1.34 trillion of agency mortgage-backed securities (MBS), with the remaining $151 billion made up of treasury inflation protected securities and agencies. That compares with a range of holdings between $473 billion and $786 billion between July 2003 and September 2008.
The post-crisis target for the Federal Reserve’s SOMA account will be contingent on many factors, but it does not appear to be unreasonable to anticipate that the central bank will ultimately have to shed between $2 and $2.5 billion of assets in its unwind.
Obviously, the first steps that the Federal Reserve will undertake will be to taper its monthly purchase of securities and then cease quantitative easing altogether. This is widely expected to begin before the end of 2013. Federal Reserve chairman Ben Bernanke has indicated that he does not expect rates to rise until mid-2015, almost two years from now. This anticipated increase in rates has been tied to a fall in the unemployment rate below 7 percent. Ignoring the possibility of accelerated economic growth, let’s focus on mid-2015 as the point where the Federal Reserve begins to withdraw excess liquidity.
The run-off of the Federal Reserve’s SOMA holdings is minimal between now and mid-2015. The treasury portfolio has $2 billion of maturities over the next 21 months, and the only other reduction will come from principal pay downs on the Federal Reserve’s MBS holdings. But what about SOMA’s maturity profile going forward? Beginning in mid-2015, the SOMA portfolio pays down $144 billion in year one, $170 billion in year two and $261 billion in year three. Contrast this with the increase of almost $900 billion in the Federal Reserve’s holdings between September 2012 and September 2013 and you will see the task that the central bank faces in unwinding quantitative easing.
While it is likely that the pace of the unwind of quantitative easing will be slower than its build up, it still appears that the Federal Reserve will have to take steps to withdraw liquidity from the financial markets at a pace that substantially exceeds the SOMA runoff.
The Securities Industry and Financial Markets Association’s 1 October 2013 published statistics reported a daily average of approximately $1.5 trillion of treasuries outstanding on repo transactions during September. The obvious question becomes how much more treasury collateral can the market absorb on a short-term basis. With markets currently trading toward the bottom of the Federal Reserve’s 0.00 to 0.25 range, there clearly is a fair amount of slack that can be taken up.
But if the Federal Reserve is faced with the need to aggressively withdraw liquidity, it may find that this task exceeds the market’s demand for overnight investments. The question that the central bank has left unanswered is whether it will also establish a programme for reducing its SOMA holdings through outright sales
Since that initial announcement, the Federal Reserve has posted a significant amount of information on its website, which, when read through, debunks a number of the more cynical interpretations of this new tool, but also leaves a bigger question unanswered.
The FRFARRF allows market participants to ‘reverse in’ collateral from the Federal Reserve, therefore investing cash and, from the Federal Reserve’s point of view, draining liquidity from the banking system. The Federal Reserve created a pilot programme to run from 23 September until 29 January 2014. By its terms, participants can bid for up to $500 million of collateral on an overnight basis (this could be increased to as high as $1 billion). Rates were set at 0.01 and could be raised to as high as 0.05. The window for these operations is between 11:15 and 11:45am EST. Only the Federal Reserve’s treasury securities are to be used.
Currently, 139 institutions are eligible to participate: at present, this includes 21 primary dealers, 18 banks, six government-sponsored enterprises (including Fannie Mae and Freddie Mac) and 94 large 2a-7 funds. This facility contrasts with the Federal Reserve’s traditional reliance on its primary dealer network exclusively to affect short-term monetary policy. The first operation was conducted on 23 September, and subsequently the Federal Reserve announced on 26 September that it would increase the maximum allocation limit to the $1 billion level.
The market can often overreact when a surprise announcement comes from the Federal Reserve. It was immediately pointed out that the central bank would now be conducting open-market operations directly with market participants at the expense of its primary dealer network. Some speculated that the creation of this new facility could partially reflect the Federal Reserve’s dissatisfaction with the pace of triparty reform. Finally, there was the prediction that the central bank’s presence in the overnight markets could raise dealers’ financing costs and serve as another lever to force them to reduce their balance sheets.
While the FRFARRF will undoubtedly alter the short-term funding markets, it is worth reading the Federal Reserve’s comments on the purpose of this facility as it is explained on its website.
The Federal Reserve makes two key points. First, the facility will complement the its payment of interest on overnight excess reserves, allowing the central bank to directly influence short-term secured and unsecured rates. Second, the facility is intended to work in conjunction with the Federal Reserve’s more traditional capped allotment repos. The central bank goes to great lengths to refer to this facility as an operational exercise, but it seems likely that an expanded FRFARRP is intended to play a major role in the Federal Reserve’s transition from its current accommodative stance to one where it will be withdrawing liquidity.
The most obvious consequence of an expanded, permanent FRFARRF would be the creation of a floor under overnight repo rates. If an institutional investor can access collateral from the Federal Reserve at 0.25 in the late morning, the investor will demand a higher rate from its market counterparties in the early am. The interesting question is how much of a premium dealers will have to pay.
While some market commentators have speculated that the Federal Reserve has abandoned its traditional reliance on its primary dealer network, an alternative explanation may be that it recognises that with all of the regulations aimed at limiting dealers’ leverage, they may not be able to play their traditional role of serving as a conduit for the Federal Reserve’s open market operations. Other considerations may be the central bank’s ongoing efforts to ensure the viability of 2a-7 funds and to compensate for collateral shortages around reporting dates.
The big winners appear be Fannie Mae and Freddie Mac, as they can now buy repos from the Federal Reserve, whereas they currently cannot access the IEOR programme. This will further reduce the volume in the overnight Federal Reserve funds market, as Fannie Mae and Freddie Mac currently have no alternative but to place their excess liquidity there.
Now, for the unanswered question: will the Federal Reserve rely exclusively on the overnight repo market to unwind quantitative easing when the time comes to withdraw liquidity?
As of the 25 September, the Federal Reserve held $3.45 trillion of securities in its System Open Market Account (SOMA) programme, $1.96 trillion of treasuries and $1.34 trillion of agency mortgage-backed securities (MBS), with the remaining $151 billion made up of treasury inflation protected securities and agencies. That compares with a range of holdings between $473 billion and $786 billion between July 2003 and September 2008.
The post-crisis target for the Federal Reserve’s SOMA account will be contingent on many factors, but it does not appear to be unreasonable to anticipate that the central bank will ultimately have to shed between $2 and $2.5 billion of assets in its unwind.
Obviously, the first steps that the Federal Reserve will undertake will be to taper its monthly purchase of securities and then cease quantitative easing altogether. This is widely expected to begin before the end of 2013. Federal Reserve chairman Ben Bernanke has indicated that he does not expect rates to rise until mid-2015, almost two years from now. This anticipated increase in rates has been tied to a fall in the unemployment rate below 7 percent. Ignoring the possibility of accelerated economic growth, let’s focus on mid-2015 as the point where the Federal Reserve begins to withdraw excess liquidity.
The run-off of the Federal Reserve’s SOMA holdings is minimal between now and mid-2015. The treasury portfolio has $2 billion of maturities over the next 21 months, and the only other reduction will come from principal pay downs on the Federal Reserve’s MBS holdings. But what about SOMA’s maturity profile going forward? Beginning in mid-2015, the SOMA portfolio pays down $144 billion in year one, $170 billion in year two and $261 billion in year three. Contrast this with the increase of almost $900 billion in the Federal Reserve’s holdings between September 2012 and September 2013 and you will see the task that the central bank faces in unwinding quantitative easing.
While it is likely that the pace of the unwind of quantitative easing will be slower than its build up, it still appears that the Federal Reserve will have to take steps to withdraw liquidity from the financial markets at a pace that substantially exceeds the SOMA runoff.
The Securities Industry and Financial Markets Association’s 1 October 2013 published statistics reported a daily average of approximately $1.5 trillion of treasuries outstanding on repo transactions during September. The obvious question becomes how much more treasury collateral can the market absorb on a short-term basis. With markets currently trading toward the bottom of the Federal Reserve’s 0.00 to 0.25 range, there clearly is a fair amount of slack that can be taken up.
But if the Federal Reserve is faced with the need to aggressively withdraw liquidity, it may find that this task exceeds the market’s demand for overnight investments. The question that the central bank has left unanswered is whether it will also establish a programme for reducing its SOMA holdings through outright sales
NO FEE, NO RISK
100% ON RETURNS If you invest in only one securities finance news source this year, make sure it is your free subscription to Securities Finance Times
100% ON RETURNS If you invest in only one securities finance news source this year, make sure it is your free subscription to Securities Finance Times
