Keynote Speech Speech at the 26th Annual Securities Finance and Collateral Management Conference
Check against delivery.
Ladies and gentlemen
Thank you very much for your kind invitation to give today’s keynote speech here at this high-level conference. For me, the board member responsible for market operations at the Bundesbank, this conference constitutes a great opportunity to exchange views on current challenges in the fields of securities finance and collateral management.
“The convergence of liquidity” is a promising title. To be honest, though: At least with regard to the liquidity the Eurosystem injects into the markets, there are – unfortunately – no signs of convergence towards a certain value yet. Most of you will be well aware that the Governing Council of the ECB decided two weeks ago to leave the Asset Purchase Programme (APP) unchanged for now. The Eurosystem will continue to buy securities through the APP to the tune of EUR 60 billiona month at least until December 2017. Hence, mathematically speaking, linear growth is the name of the game – at least for now. Excess liquidity, an essential indicator for monetary policy implementation, currently stands at EUR 1,650 billion.
I would like to take this opportunity today to elaborate on three issues that are, at least to some degree, interlinked with the Eurosystem’s activities in the markets. First, I would like to discuss the remarkable upswing in asset prices after the financial crisis and drivers for this development. Second, I would like to briefly discuss the shifts in financial markets with respect to market liquidity. I will argue that, whereas some of the factors affecting market liquidity are temporary, others – such as the regulatory environment and innovations in trading – are here to stay. Finally I will conclude by presenting my view on the outlook for monetary policy implementation.
2 Current drivers of financial markets
The financial crisis of 2008 and the subsequent European debt crisis dealt financial markets a serious blow. Investors had to reconsider long-held beliefs that were accepted as truths: that large banks were immune to bankruptcy, and that sovereigns in the developed world would never default on their debt. When the crisis hit and the unimaginable happened, prices of financial assets, apart from a few viewed as safe havens, went into a tailspin.
This is history. Asset prices have rebounded remarkably. Stock markets in Europe as well as the US have reached new historic heights. Corporate debt is in high demand, and risk premia have shrunk noticeably, even for heavily indebted entities. Prices have also increased for more tangible assets such as real estate. Naturally, the continued rise in asset prices is fuelling speculation about potential bubbles. However, judging whether price developments constitute a bubble or reflect fundamental and sustainable factors is not a simple task.
2.1 Recovery in asset prices and QE
There is little doubt that the extraordinary measures taken by central banks to shore up liquidity have played a significant part in these price developments. They reassured markets, rebuilt confidence and created an environment in which investors were once again willing to put money into risky assets. It is important to remember that this was an intended effect. For the economy to grow again, businesses and households need to have access to affordable capital. Investment is the lifeblood of any recovery.
However, there is, of course, the concept of too much of a good thing. Unintended side effects of quantitative easing are becoming more apparent the longer the programme runs and the less it is needed to sustain the economic recovery. So far, the positive effects may well have outweighed the negative implications, but this is bound to change at some point.
Quantitative easing is designed to work by nudging investors away from safe havens and towards riskier assets. However, one has to realise that the balance to be struck is rather delicate. There is a serious risk that investors will start to underestimate the risks of their investments, leading to price distortions and misallocation of capital. An abundance of liquidity will not by itself make a risky investment less so. Market participants should think very carefully about what will happen once the ample liquidity provision stops.
Additionally, there is the thorny issue of whether private investors are being crowded out of certain market segments by central banks and what the long-term effects are. Although the Eurosystem puts a high value on implementing the purchase programme in a market-neutral way, it must be acknowledged that there are areas of concern.
Apart from concerns about private investors simply being crowded out for the time being and associated questions about market functioning, there is also concern about which investors will come back once the purchase programmes end. It is certainly not the Eurosystem’s intention to harm capital markets: quite the contrary. Nevertheless, it needs to be recognised that investors probably will most likely not flood in again immediately.
But as a wise person once said: it is very difficult to predict – especially the future. Instead of looking into a crystal ball, I would like to discuss two further drivers that play a major role for asset valuations.
2.2 The macroeconomic environment
The first one, the current macroeconomic environment, is convincing. While the recovery was relatively anaemic for a long time, it seems to have gathered pace over the last two to three years. This is in part thanks to the unconventional measures taken by the Eurosystem.
Quarterly growth in the euro area has improved and has ranged between 0.2 percent and 0.8 percent for the last 16 quarters. The latest reading for April puts the unemployment rate at 9.3 percent. That is still too high, but it is nevertheless a marked improvement on its peak of 12.1 percent in mid-2013. And spirits are up: economic sentiment is close to a nearly 10-year high.
Inflation, however, remains a sore point. Accelerating inflation from the beginning of the year has come down once again. After hitting 2.0 percent in February, its highest reading since January 2013, inflation has retreated to 1.4 percent in May. The oil-price effect has subsided and once again the onus is on economic growth to generate inflation. The Governing Council of the ECB puts a strong emphasis on underlying inflation. It is very important for credibility that policy decisions are taken on a strong footing. They cannot be beholden to monthly volatility in data publications.
So overall, the macroeconomic backdrop does provide some justification for price developments over the last few years. Asset prices should rise when growth prospects improve. Naturally, it would be welcome to see asset price inflation translated into a reasonable level of consumer price inflation as well.
2.3 Political uncertainty
While the macroeconomic backdrop is promising, there are still complications ahead. Politics, another driver for asset prices I would like to briefly discuss, has created a lot of headlines over the last year. These developments have not substantially affected the high valuation levels, though.
The first major test came in the form of the Brexit referendum in June 2016. The decision by a majority of UK voters to leave the European Union came as quite a shock. Markets exhibited substantial volatility in response. This response was short-lived, though, and markets have remained orderly throughout. Next up was the presidential election in the United States. Markets adjusted even more quickly to the new circumstances. The constitutional referendum in Italy in December 2016 was barely registered in markets.
However, political risks remain a concern for markets. The run-up to the French presidential election had investors worried, as evidenced by a widening spread between French and German sovereign bonds. And the view on the US has somewhat turned from exuberance to wary idealism as it has become clear that implementing all the stimulus measures promised by the Trump Administration will not be a walk in the park after all.
Looking forward, questions about the US political agenda will remain relevant, and the road ahead of the Brexit negotiations will give investors much to interpret and think about. The outcome of the UK elections might very well complicate the negotiations with the rest of the European Union further. In addition, election season in Europe is not quite over, with the German general elections due in September.
The Brexit negotiations will not only be reflected in the prices on financial markets, but will also have a lasting effect on financial market locations themselves. Once the UK has left the European Union, European agencies currently based in the UK, like the EBA and EMA, will have to be relocated into the EU-27. The future location of euro clearing operations is certainly another crucial aspect: in my point of view, euro clearing is of key importance for Europe, and a mere supervisory recognition of a UK central counterparty as “equivalent” is unlikely to be sufficient post-Brexit.
Banks and the financial services industry will need to waste no time in planning ahead for the future, since Brexit could involve a substantial loss of rights to conduct business in the European Union and possibly necessitate the relocation of sensitive and important infrastructures. That is why I expect a growing number of firms to transfer their operations from the UK to continental Europe even before negotiations have been finalised. There will not only be just one city which attracts these companies. Cities such as Paris, Dublin, Luxembourg and, of course, Frankfurt will complement one another in terms of the value they add. And, in my personal opinion, Frankfurt has quite a few major selling points to offer.
Looking at current asset valuations, markets appear to be taking the political risks in stride. Going forward, this behaviour might very well continue with markets only adjusting once it looks like politics will impact on the fundamental outlook. That is what is happening right now with the market participants assessing the reform agenda in the US, and probably will happen when the Brexit negotiations are underway. If some tail risk scenario were to materialise, however, markets might reassess valuations a lot more forcefully, and prices of risky assets might then adjust accordingly.
3 Transitory and structural changes in market liquidity
Ladies and gentlemen
Distortions in asset prices are one concern when discussing financial markets. Other concerns are related to changes in the market place. Today, we are facing a situation where central bank liquidity is hitting all-time highs daily while, at the same time, market liquidity appears to be fickle in a number of segments. Reason enough, in my view, to shed some light on factors that influence market liquidity.
3.1 The APP and securities lending
As already mentioned, the purchase programmes seem to have been successful in lowering fixed income yields and fighting potential deflation risks. Nevertheless, they have been criticised strongly. Not only for inducing concerns in matters of supporting further government debt and delaying structural reforms, but also because they are said to create market distortions.
Academic evidence on the effects of non-standard measures is helpful for determining the need for remedial measures, but unfortunately rather rare. The Bundesbank and the Bank for International Settlements have launched a joint project which analyses the impact of purchases within the Public Sector Purchase Programme (PSPP) on prices and liquidity of Bunds in the first 18 months of the programme. The researchers provide evidence for negative side effects over the lifetime of the programme, eg decreasing Bund market liquidity. Based on data from the Bund cash market, the authors not only show deteriorating liquidity conditions over the course of the programme, but also find evidence that the purchases have a negative effect on transaction costs and order book depth. These findings suggest that the scarcity effects of asset purchases can impact adversely on market functioning.
Strains in market liquidity have also become visible in the repo market. Especially at quarter or year ends we have seen deeply negative repo rates. This can be attributed to a variety of reasons, but the large-scale purchases of the Eurosystem may have played a role.
In order to curtail tensions in repo and bond market liquidity, Eurosystem central banks have made their holdings available for securities lending from the start of the PSPP. Let me go a little deeper here, as securities lending is a key topic at this conference.
The lending was implemented in a decentralised manner in the Eurosystem. The Bundesbank started by using the automated Clearstream lending facilities. In September 2016 we also began to offer a bilateral repo facility.
Not all the features of the bilateral repo facility have been fully appreciated by market participants. We have imposed relatively strict operational and risk management parameters. This concerns specific terms such as a maximum maturity of seven days or the exclusion of automatic rollovers as well as certain collateral requirements.
I would like to stress that the lending facilities for the asset purchase programmes are designed as backstop facilities. We do not want to crowd out private players; we are just offering supplementary facilities to the repo and lending markets. Against this background, we regard somewhat stricter operational parameters as justified. Furthermore, as a central bank we have to place particular emphasis on risk mitigation, and most of the lending parameters serve exactly this purpose.
Despite the criticism put forward by some market participants, in the meantime our repo facility appears to be being well received in the market. So, allow me to note that the conditions do not appear to be overly harsh after all. The outstanding volume has been continuously growing and most of our active trading counterparties are also using the repo facility. Especially the opportunity to submit cash collateral for PSPP securities lending, which was implemented in December 2016, has enhanced the popularity of the bilateral facility. The APP securities lending primarily addresses the most active market participants in the respective markets or primary dealers. It is, however, open to other counterparties as well.
In 2017, volatility in the repo market for German government securities has been rather moderate so far, even at the end of the first quarter. It seems that market participants have found a way to cope with the challenging environment. The Eurosystem’s lending facilities may have contributed to this development.
Another area in which various commentators and academics have recently criticised the Eurosystem is the collateral framework. The Eurosystem conducts repo operations to provide temporary liquidity to the banking sector and takes financial assets as collateral in return. The collateral framework defines the eligible financial assets, the related pricing methodology and the respective risk-control measures.
Critics contend that the Eurosystem is accepting collateral of insufficient quality at excessively lenient terms. This – so they argue – could create market distortions and large risks for the Eurosystem’s balance sheet.
The Eurosystem is taking this criticism seriously. However, it should be duly noted that, throughout the difficult times of the financial and sovereign debt crisis, the broad collateral framework was regarded as a great institutional strength of Eurosystem’s monetary policy. While the pledged collateral has successfully protected the Eurosystem from financial losses, at the same time the collateral framework has been flexible enough to preserve the Eurosystem’s ability to support market functioning and provide liquidity to the banking system in times of stress. The Eurosystem thereby fulfilled one of the core functions of a central bank.
As is apparent from the agenda of this conference, regulatory issues have become a crucial factor in financial markets, cropping up in almost every agenda item. In the aftermath of the financial crisis, national and European regulatory initiatives alike have mushroomed.
In a nutshell, the regulations involve stricter rules for capital, stricter provisions for risky transactions, and higher administrative burdens. At the Bundesbank, we are fully aware of these strains and the potential drawbacks of regulatory requirements. Liquidity in the bond and repo markets has decreased, also – but not only – because of new regulations.
Moreover, the higher required capital ratios will increase banks’ cost of capital while banks’ earnings are currently compressed. This is making it difficult for banks to raise further capital, which is why they may respond to regulatory requirements by deleveraging. This, in turn, may well have negative effects on market liquidity.
At the same time, it is evident that well-designed regulatory measures will help make the financial system more stable. The ultimate aim is to make financial markets more resilient and crisis-proof in the future. Higher resilience will restore market and customer confidence in banks. This is intended partly make up for the higher costs of capital and allow banks to attract more customers and funds. In the long run, a stronger capital base can be expected to enhance banks’ loan origination capabilities. These are thus positive effects of regulation which need to be recognised.
I am convinced that we need effective and well-targeted regulation. This also means that we need to closely monitor all effects – intended and unintended – and consider making adjustments, if necessary.
But even if targeted adjustments are made to the regulatory framework, I believe one thing can be expected: while other factors currently weighing on market activity are eventually going to fade out, tighter regulatory requirements will definitely stay, and this is the “new normal” market participants will have to adapt to.
3.3 High Frequency Trading
Let me now turn to another phenomenon whose role in market liquidity is, in my view, larger than some observers think. For the last fifteen years the investment community has witnessed a transformation in the landscape of exchanges, brokers and investors from a trading process that is based on human decision making into an automated electronic high-speed exchange. High Frequency Trading (HFT) operates in a matter of microseconds.
The Bundesbank conducted a comprehensive empirical study on the significance of HFT in the German capital market and analysed whether or not HFT has had a positive impact on market liquidity.
The results show that HFT with its 40 percent share of trading volume in the DAX and BUND future, does indeed play a very dominant role in market making and in the supply of liquidity in these very important market segments. The observed characteristics of the market behaviour paint a two-sided picture of HFTs’ impact in the markets:
In normal trading environments, the findings support the view that many HFTs act as a stable source of liquidity that provides investors a steady opportunity to trade future contracts at very good prices. Market response times to incoming fundamental news are in the millisecond range and bid-ask spreads are for the most part at their minimum.
During times of increased stress in the markets, such as during the anticipated announcement of important macroeconomic data, the empirical findings show a change in HFTs’ behaviour. Market-making activity of high-speed algorithms is significantly reduced, sometimes close to zero for short time periods, while at the same time more aggressive and liquidity-demanding HFT trading strategies take over and start to dominate the trading field.
Hence, the ramifications of HFT and the transformation into an automated trading ecosystem indeed change the liquidity conditions in the market. Drawing the right conclusions is not easy – for market participants and regulators alike.
4 Outlook for monetary policy implementation
To conclude, there are a number of factors which impact financial markets. While the technological progress and the regulatory changes are here to stay, the Eurosystem’s extremely accommodative monetary policy will come to an end at some point.
Discussions on normalisation took centre stage again last week, when the US Federal Reserve System updated plans for its balance sheet runoff. The duration of the process has been defined as “a few years”. The general approach will be to let the balance sheet runoff happen passively “in the background,” as if markets were simply “watching paint dry.”
It is self-evident that the Eurosystem is hoping for a process which will be just as boring as the Fed’s plan. But first, the necessary precondition has to be met:inflation has to sustainably adjust back towards the ECB Governing Council’s target for price stability of below but close to 2 percent in the medium term. Additionally, we must bear in mind that – like in the US – the exit process will take several years.
Which technical options does the Eurosystem have at its disposal once the Governing Council determines that it is time to start the exit process?
From a technical point of view, neither an increase in the level of key interest rates nor an exit from individual non-standard monetary policy measures would pose particular challenges. However, since the start of the financial crisis the Eurosystem has gradually implemented a large number of expansionary measures which would need to be phased out. Once the Governing Council has determined the appropriate timing and has adjusted its “Forward Guidance” accordingly, the sequencing of monetary policy normalisation and exit from non-standard measures will follow a gradual and prudent approach. Even after non-standard measures have been discontinued, a considerable amount of liquidity will stay in the market for a prolonged period.
For the Bundesbank, the key step towards the normalisation of monetary policy would be the end of outright purchases of government bonds under the Asset Purchase Programme, as this programme blurs the distinction between monetary and fiscal policy. However, from an operational or economic perspective there is no predetermined sequence for exiting from single non-standard measures. It should be noted that the comprehensive package of non-standard measures is highly expansionary. Consequently, removing one measure would only remove its particular effects, i.e. the monetary policy stance would not become restrictive, only less expansionary.
Hence, a number of different options for the departure from expansionary monetary policy measures are possible. An orderly exit which has been properly prepared both technically and verbally will allow a smooth transition in the financial markets. Talking about communication – a look at the introductory statement of the press conference after the last ECB Governing Council meeting reveals a few interesting details. The Governing Council considers that risks to the growth outlook are now broadly balanced. It also removed the easing bias on rates for now. Some observers interpreted this statement as a further step towards monetary policy normalisation – and market participants should be prepared for these new challenges.
Thank you for your attention.