## Paul Fisher: The corporate sector and the Bank of England's asset purchases
Speech by Mr Paul Fisher, Executive Director for Markets and Member of the Monetary Policy Committee of the Bank of England, for the Association of Corporate Treasurers Winter Paper 2010, London, 18 February 2010.
The original speech, which contains various links to the documents mentioned, can be found on the Bank of England's website.
* * *
## The start of corporate sector asset purchases
On 19 January 2009, the Chancellor announced 1 that the Bank of England was being authorised to purchase 'high quality private sector assets', financed by the issue of Treasury Bills '… to increase the availability of corporate credit, by reducing the illiquidity of the underlying instruments …'. In this talk today, I want to address the rationale for intervening in this way and the opportunities and limitations of such operations. I will then go on to review the individual schemes and evaluate the results so far. I will not be talking very much about the wider programme of gilt purchases, which had different objectives - namely to directly boost the money supply and nominal demand. But there are some interactions which I shall come back to later.
At the time the policy was announced, output in the United Kingdom economy was dropping sharply. GDP has fallen by 6% during the recession overall and fell by 2 ½% in 2009 Q1 alone, at the time these policies were being put in place (Chart 1). That is the largest recorded quarterly fall in output since 1958. In the wake of the financial dislocation of Autumn 2008, there was a collapse of confidence and elevated uncertainty which had a profound effect on consumer demand and investment intentions. There was also an intensification of the 'Credit Crunch' which had been in train since the start of the crisis in mid-2007.
A range of policies had been introduced in the United Kingdom to prevent an economic meltdown. That included action to re-capitalise some of the banks; guarantee some of their funding; expand liquidity assistance and to set appropriately accommodative fiscal and monetary policies. Governments around the world took very similar steps. Asset purchases were one part of the policy response.
At its peak, the Bank of England bought nearly £3bn of commercial paper and corporate bonds (Chart 2). This is sometimes compared with the eventual total of £198bn gilts purchased, prompting the question why the Bank didn't buy more private sector assets? This question reflects a - perhaps natural - misunderstanding of the different purposes of these interventions. The gilt purchases were intended as a monetary policy operation to inject a substantial amount of money into the economy. The consequences of that policy have been addressed in Bank publications (Bank of England, Quarterly Bulletin, June 2009 for example), speeches and elsewhere. There has been much less discussion about the private sector asset purchases. So to address that topic tonight, I would like to first set out some principles, as I see them, behind the actions of central banks in so called 'credit-easing' operations.
1 The announcement is at http://www.hm-treasury.gov.uk/press\_05\_09.htm. Subsequent letters exchanged between the Governor and Chancellor were published on 29th January and can be found at http://www.bankofengland.co.uk/publications/news/2009/005 and those published on 5th March can be found via http://www.bankofengland.co.uk/publications/news/2009/019.htm. The letters published on 5th March were dated 17th February, from the Governor, and 3rd March from the Chancellor.
## Intervention by the authorities
The Bank of England intervenes routinely in financial markets so as to implement monetary policy and provide liquidity insurance to the banking system.
These operations, normally restricted to banks and building society counterparties, stem directly from the Bank's core responsibilities of maintaining monetary and financial stability. But these responsibilities do not give the Bank a mandate or the ability to provide a source of long-term funding for the commercial banking system 2 , nor is it for the Bank to take material risks with public money. In the case of monetary policy, the central bank expands and contracts its balance sheet to meet its objectives. Typically that involves injecting and/or draining cash (central bank money) in order to set interest rates or, as with quantitative easing programmes, changing the money supply directly. The Bank also provides liquidity support to commercial banks in times of stress (for example through its Discount Window Facility). But such lending is well-collateralised, so low risk, and is always intended to be temporary: to overcome a one-off shock or to bridge to a more lasting solution. To discourage reliance on these operations, standard practice is to charge a penalty rate relative to market prices. These liquidity insurance operations are an integral part of the Sterling Monetary Framework which delivers our monetary policy objectives (See Fisher, 2009 for further details).
The Bank of England does not have access to funds that could be used to sustain commercial lending operations. We do not have the power to raise tax, nor do we have significant retail deposits, nor the freedom to borrow in large scale in the wholesale markets for non-monetary purposes. 3 More generally, in a capitalist economy, it is private sector or, in some cases, national savings which ultimately fund the banking system. And it is the job of financial intermediaries to translate those savings efficiently into private sector spending.
In contrast, the fiscal authorities (including national wealth funds where they exist) can access private sector funding via taxation or by borrowing in large scale. There are many instances throughout UK economic history where the government has intervened to support private markets, or indeed individual firms. There are well-known risks associated with such action, including the loss of public money and/or an inefficient allocation of capital in the economy. There are also substantial and well-known potential benefits if the intervention corrects a market failure or achieves a wider social objective. The merits of such interventions need to be judged on a case-by-case basis. For tonight's discussion I just want to argue that such decisions should be undertaken with the support of the democratic political system and should be implemented as a fiscal operation. Indeed, during this crisis, the UK Government has made a number of such interventions - such as the Enterprise Finance Guarantee Scheme and the Vehicle Discount or 'Scrappage' Scheme.
## The Special Liquidity Scheme
Now, as an aside, some keen watchers of central banks may want to argue that the Bank of England's Special Liquidity Scheme (SLS) appears to have provided £185bn funding for the UK banking sector. The SLS, supported by a Government indemnity, was probably the single most generous liquidity support scheme introduced by a central bank during the crisis. 4 Its
2 In the past, when financial markets were less well developed, the Bank of England has played a part in selected industrial or financial market interventions (see Sayers, 1976), but that is not a role we have now.
3 The Bank of England issues banknotes of course, but these are monetary liabilities or 'central bank money' which impact on the monetary stance. They are backed only by the highest quality assets, in order to maintain confidence in the integrity of the currency.
4 This was required given the relatively small capital base of the Bank of England - around £3 ½ bn at end-Feb 2009.
detailed characteristics serve to make several points in the context of this talk. The SLS did not provide the banks directly with funding. Rather it swapped illiquid private sector assets held by the banks for highly liquid treasury bills borrowed from the Debt Management Office. The commercial banks then obtained their funding from the market by repo of the treasury bills. Essentially the SLS provided a temporary collateral upgrade for the banks. By the time the scheme closes, SLS participants need to have either generated alternative sources of funding or created alternative collateral which the market will be willing to hold in place of the treasury bills.
Recently, there have been some calls from interested parties for the SLS to be extended beyond its three-year horizon. If the SLS were to be extended, that would mean an implicit subsidy for specific banks. Albeit for a fee, those banks would be funding on an ongoing basis their illiquid, risky assets, by borrowing at 'risk-free' government interest rates. That would give a significant competitive advantage to banks that had most recourse to the SLS. But central bank operations are not carried out to protect individual banking franchises, they are there to ensure overall monetary and financial stability.
As the Governor has already made clear, the SLS will close at the end of January 2012. The Banks liquidity insurance facilities are, and will be available - but, as I have noted, these are for temporary usage during periods of stress, are set at an appropriate price and are not there to provide sustained funding for the banking system.
## Market-making operations by the central bank
As part of its broader remit, however, a central bank is fundamentally concerned with maintaining conditions for the stable provision of financial services to the wider economy. This crisis has demonstrated that specific financial markets can sometimes fail and that raises the possibility of the central bank stepping in as 'Market Maker of Last Resort' (MMLR).
Acting as a MMLR is not something of which we - or other central banks - have had a lot of experience, but there are some basic principles which, if followed, can allow such operations to be run by the central bank consistent with its other objectives. My colleague, Paul Tucker, outlined these in a speech last year (Tucker 2009). The most relevant are:
- (i) MMLR activity should not interfere with the setting of monetary policy by creating additional central bank money independent of monetary policy operations. 5
- (ii) Transactions should be at a discount to fundamental values so that the operations are unattractive to the markets in normal conditions.
- (iii) The operations should help to re-start, or energise markets rather than replace them.
These principles all apply to the Bank's corporate asset purchases. In fact, when the Asset Purchase Facility (APF) was first introduced, the Bank was effectively operating as an agent for the Government: the purchases were financed by the issue of treasury bills, not by the Bank, and the Government indemnified the Bank against any losses. Had the Government wanted to, the purchases could have been designed and operated as a fiscal operation and not implemented by the Bank. But the correspondence from the Chancellor makes it clear that the purchase of private sector assets (i) was intended to improve the functioning of markets and so was consistent with a MMLR function and (ii) it simultaneously provided a framework for asset purchases as a monetary policy operation by the MPC and so the Facility had to be designed to be compatible with that.
5 Once the programme of quantitative easing was established, the corporate purchases were funded by creating central bank reserves. But that was consistent with implementing the MPC's monetary policy.
## Motivating the APF schemes
The Bank's initial APF interventions focussed on sterling corporate bonds and commercial paper. These markets are relatively small. At the end of 2006, before the onset of the crisis, the outstanding stock of sterling non-financial commercial paper was around £5bn, and sterling non-financial corporate bonds totalled around £70bn. That is low compared with the stock of commercial bank lending (lending by UK resident banks to UK non-financial corporates was £437bn at end 2006). But despite their relatively small size, these markets are vitally important.
## Commercial paper
Commercial paper (CP) is a common way for larger firms to manage their day-to-day cash needs. And insufficient cash flow is one of the key ingredients in corporate failures. At the start of 2009, commercial paper spreads were way above any estimate consistent with the risk involved: they implied default rates around 50% higher than witnessed in the Great Depression. This market seemed like a prime candidate in which the APF should intervene and so facilitate otherwise credit-worthy firms in maintaining their access to short-term finance.
In trying to catalyse a market, the Bank could not operate like a private investor or bank which would make purely discretionary judgements about whose CP to buy. Rather, a facility was set up which, in principle, any firm could access if they met fairly broad criteria. The APF offered to buy up to three-month CP at spreads which were significantly below those in the market at the time, but which were significantly above those expected to prevail in normal conditions. Initially this would help to drive market spreads down and the APF would then revert to providing a 'backstop' offer. In this way the scheme was set up to be self-liquidating as normal market conditions returned.
There were some important restrictions. Most notably, only broadly investment grade paper could be bought, consistent with the terms initially set out by the Chancellor and so limiting the degree of credit risk taken. Eligible firms also had to contribute materially to the UK economy and must not be banks (other policies, as noted, had been introduced to help banks). And individual counterparty limits were imposed.
Although disentangling the precise impact of the scheme from the global improvement in financial markets is difficult, it does appears to have been successful. Purchases peaked at around £2.4bn in late April 2009, at that time amounting to around one third of the sterling CP market. But, as I have stressed, the objective of the scheme was not to purchase a large quantity for its own sake. It was much more important that market spreads for the average issuer started to fall soon after the scheme was launched. Market spreads for all A1 rated sterling paper fell below APF spreads in August 2009 (Chart 3). By February 2010 nearly all firms who had sold CP to the APF were able to re-issue at cheaper rates in the market and the Bank's CP holdings have fallen back to under £300mn. That decline in usage, together with greater ability to issue in the market, indicates the improvement in market conditions which the scheme was intended to help bring about. I should note that, at the current moment, the need for firms to issue CP is at something of a low point. Investment-grade firms do not seem to have a strong demand for short-term cash and are concentrating on issuing longer-term debt. We have promised to keep the CP scheme under review and give 12 months notice before it is withdrawn.
## Corporate bonds
A corporate bond scheme was also set up last year. In January 2009, the primary market was open for new bond issuance, but spreads over gilts were extremely wide, and were widening further as issuance increased (Chart 4). It was clear there was a substantial
liquidity premium in the market. Secondary market liquidity was notoriously poor for sterling denominated bonds. The small size of the bond market, relative to euro- and dollardenominated markets, meant that many individual sterling-denominated bonds might not trade for months at a time. There was a risk that secondary market illiquidity might have restricted the strength of the primary market, so constraining the availability of credit.
Most corporate bonds are bought by asset managers for long-term investment, but investors do need to actively manage their portfolios e.g. for duration and credit risk. The scheme was designed to meet this need by offering to buy small amounts of bonds in the secondary market on a regular basis. There were two immediate aims. First, by regularly posting traded prices, it was hoped to improve the process of price discovery and hence aid secondary market liquidity. Second, by offering a bid in the secondary market, the scheme should encourage further private sector demand in the primary market. That leaves open the question of whether we could or should have been trying to buy much larger quantities. Had the APF done so, in addition to strong primary market private sector demand, it would undoubtedly have helped to drive down market spreads. But it would also have risked displacing that private sector demand, to little purpose in my view. In fact, a greater intervention in this market during this period could have made things worse - the problem was in the functioning of the market, not a lack of demand for the assets.
It would be fair to say that the corporate bond scheme attracted some criticism in its early months. Many observers misunderstood the nature and purpose of the scheme and considered that the small amounts bought were a sign of failure. But, as I have stressed tonight, the objectives of the scheme were not based on buying a particular amount. In fact, over the past year, the corporate bond market overall has performed consistently strongly and in line with the original objectives of the scheme: There has been record issuance of sterling corporate bonds (Chart 5); spreads for eligible bonds have halved (Chart 6); bid-offer spreads have narrowed; and the bond-CDS basis 6 has also contracted significantly. These asset purchases do seem to have played their part in improving the access of investment grade corporates to bond market financing.
The one disappointing feature of the sterling corporate bond market is the continuing problem of secondary market illiquidity. Market-making dealers have not been able to allocate much capital to holding inventories of bonds and hence the market is still not functioning as well as it might. This was one reason why, in January this year, the Bank introduced operations to sell as well as to buy bonds. Further market evolution may be needed to overcome this illiquidity. One possible way forward would be more trading on exchanges. It will be interesting to see whether the recent listing of bonds on the LSE in retail-size amounts is successful.
## Secured commercial paper and supply chain finance
The Bank also considered intervening in other markets such as secured commercial paper (SCP), supply chain finance, syndicated loans and more broadly in ABS markets, as set out in the Chancellor's original letter to the Bank. 7
Taking secured commercial paper first, this is a form of asset-backed commercial paper (ABCP) based on a bundle of relatively short-term loans to a variety of borrowers. Due to over-collateralisation, the security might be investment grade even if the underlying borrowers are not. By catalysing this market, smaller or lower-rated-firms could gain
6 The difference in price between a cash bond and the Credit Default Swap can be interpreted as a measure of the liquidity premium in cash bonds.
7 The Bank also made provision to buy securities issued under the Credit Guarantee Scheme, using the same format as for its corporate bond purchases. To date, it has not been deemed necessary to activate this facility.
improved access to credit. But, in designing a facility, the Bank was concerned not to encourage the sort of risk-taking which had caused so many problems leading up to the crisis. In particular that meant a restriction on the degree of maturity mismatch between the underlying assets and the securities they backed.
A purchase scheme was launched last July, but the Bank has not been offered any SCP to buy so far. New programmes take 3-6 months to set up and meanwhile, according to our contacts, the market for such paper, particularly in the United States, quickly improved to the point where our 'backstop bid' was no longer attractive. And we are told there has been sufficient capacity within commercial banks' existing programmes to fund the assets they were generating. Nevertheless, the facility remains open and the Bank is still in discussions with potential issuers.
Supply chain finance (SCF) is another route by which smaller firms can potentially get access to credit against high-quality assets. Typically, this would arise when a large firm has a number of suppliers who wish to raise immediate finance against their invoices. Unfortunately there was no SCF market standard and so there was not a viable market that the APF could attempt to catalyse. Last Autumn, a working group of market participants (borrowers, lenders, banks etc) gathered together to develop some proposed standards and try to build a sustainable market in this space. That group is being chaired by Stuart Siddall, Chief Executive of the ACT, and is due to issue a report in the next few months.
## Syndicated loans and asset backed securities
A substantial part of lending to larger UK corporates is normally undertaken via syndicated loans. The Bank investigated whether there was a market failure or illiquidity premium in this market which could be addressed by making APF purchases. After a thorough investigation, the Bank concluded that what was needed was more large-scale bank funding, particularly to replace the foreign bank lenders who had been withdrawing from the UK market. But the appropriate way to address that was through the various schemes in place to support the banks.
Similarly, the Bank investigated the state of securitisation markets. Here, there were two problems. First, many investors were extremely wary of buying mortgage-backed securities and similar products, because the crisis had raised uncertainty over their underlying risk characteristics. At the same time there was a huge outstanding stock of such assets already issued, which would swamp any operation aimed at improving liquidity. In the United States the authorities have undertaken substantial purchases of mortgage-backed securities, and funded the purchase of other securitised assets. It is not clear yet whether those markets will be self-sustainable once official intervention comes to an end.
In the United Kingdom, we have made a range of ABS, including mortgage backed securities, eligible collateral in the Bank's permanent liquidity insurance facilities as well as in the SLS. And the Bank intends to consult the market shortly on what the requirements should be in respect of loan-by-loan information for such assets to be eligible in the future. Along with other monetary authorities taking similar actions, 8 standardisation of information provision may help improve investor confidence in the underlying securities.
## Non-bank lending
Also during 2009, the Bank used its market and corporate contacts to identify why bank lending dominated the supply of credit and what the obstacles were to lending from nonbanks. There appeared to be no easy wins but a recent HMT consultation paper which draws
8 See the ECB announcement: http://www.ecb.europa.eu/press/pr/date/2009/html/pr091223.en.html.
BIS Review 23/2010
heavily on the Bank's investigation, explores options to facilitate the supply of credit from non-bank lenders. 9
## Interactions with other monetary policy actions
The commercial paper and corporate bond schemes seem to have been reasonably successful in helping to improve key markets and hence facilitating access to credit, at least for larger corporates. Smaller firms who supply those larger firms, may also have seen some 'trickle down' benefits. One indicator of this is that the rise in the level of company insolvencies during the recession has been surprisingly small so far, given the fall in output (Chart 7). But the major supportive factor has probably come from the very low level of Bank Rate and the programme of quantitative easing. The big moves in asset prices during 2009 - a 50% rise in equities, 150bp fall in LIBOR spreads, the 150 bps fall in targeted corporate bond spreads for example - will have been strongly supported by the Bank's monetary policy operations. And by purchasing the gilts which asset managers and others would normally have bought, the impact of the quantitative easing programme will have been to increase the demand for more risky assets - like corporate bonds. So the corporate and gilt purchase programmes have been complementary.
## Conclusion
In my talk this evening, I hope I have explained a bit more clearly the motivation for and the design of the APF corporate purchase schemes. The Bank has not sought to provide large scale funding directly using public money. Rather, the operations have been consistent with a central bank acting as 'Market Maker of Last Resort'. The Bank has transacted in modest quantities designed to improve the liquidity of corporate credit markets, bring down the cost of credit and encourage private sector lending. That has been supported by the much larger programme of quantitative easing and the low level of Bank Rate. The evidence suggests that, in conjunction with a very large monetary stimulus, abroad as well as at home, the corporate purchases have been successful in helping to invigorate those corporate credit markets in which the Bank has intervened.
## References
Benford J, Berry S, Nikolov K, and Young, C (2009), 'Quantitative Easing', Bank of England Quarterly Bulletin 2009 Q2, available at http://www.bankofengland.co.uk/publications/quarterlybulletin/qb090201.pdf.
Fisher P G F (2009), 'The Bank of England's Balance Sheet: Monetary Policy and Liquidity Provision during the Financial crisis' speech available at http://www.bankofengland.co.uk/publications/speeches/2009/speech413.pdf.
Tucker, P M W (2009), 'The Repertoire of official sector interventions in the financial system: last resort lending, market-making, and capital', speech, available at http://www.bankofengland.co.uk/publications/speeches/2009/speech390.pdf.
Sayers, R.S., (1976) 'The Bank of England 1891-1944', Volumes 1 and 2, Cambridge University Press.
9 See: 'Discussion paper on non-bank lending', HMT published on the 12 January, available at http://www.hm-treasury.gov.uk/d/non\_bank\_lending\_discussionpaper.pdf.
## Chart 1: GDP"
Chart 3: A1/P1 sterling commercial paper spreads
2009
(a) Spread of quoted yields on A-1/P1 rated UK corporate and non-bank financial sterling 3 month commercial paper over 3 month ovemight index swap rates.
## Chart 2: APF corporate purchases"
(a) Corporate bond purchases are net of sales
## Chart 4: Investment grade corporate bond spreads"
(a) Index of spread of investment grade corporate bond yields over maturity-matched risk-free rates. Source: Merrill Lynch
Chart 5: Cumulative gross issuance of sterling bonds by UK PNFCs
Source: Dealogic and Bank calculations
## Chart 7: Company liquidations in England and Wales and GDP
(a) A recession is defined as two consecutive quarters of falling output estimated using the latest data. The recessions are assumed to end once output began to rise.
- (b) Chained-volume measure at market prices.
(c) Changes to legislation, data sources and methods of compilation mean the statistics should not be treated as a continuous and consistent time series. Since the Enterprise Act 2002, a number of administrations have subsequently converted to creditors' voluntary liquidations. These liquidations are excluded from the headline figures published by The Insolvency Service and excluded from the chart.
Chart 6: Median 'APF eligible' corporate bond spreads
(a)Median of asset-swapped spread of individual APF-eligible bonds. Sources: UBS Delta and Bank calculations.