BIS Quarterly Review, September 2016 - media briefing
18 September 2016
In the quarter under review, markets struck some dissonant notes - in many ways familiar ones, but in others less so.
It all started with a huge surprise that wrong-footed market participants: on 23 June, the United Kingdom voted to leave the European Union. During the next couple of days, equity prices plunged, volatility spiked, credit spreads widened and sterling sank to a 30-year low. The markets undid, with interest, the gains they had made during the previous week, when a "remain" outcome had appeared all but certain - in market prices, not in the polls. That said, through the turbulence markets continued to function well, and market liquidity - despite all the worries - held up, with no signs of gapping. Moreover, a couple of days later, the turbulence had gone just as quickly as it had arrived. To be sure, in contrast to other big "shocks", its timing was entirely predictable: both market participants and the authorities had plenty of forewarning to prepare. And central banks were quick to provide reassurances, with both individual and joint statements that underlined their readiness to provide support to markets, institutions and the economy, if the need arose. Even so, the speed of the recovery took many by surprise, given the political and economic uncertainty that the vote had triggered.
With the UK referendum behind them, markets generally resumed the rally that had begun following the much more intense turbulence earlier in the year. By early September, equity prices in the United States were hitting new highs, those in Europe and Japan had made considerable gains since their year lows, volatilities had fallen towards post-crisis averages - or, in the case of the all-important VIX, even lower - and credit spreads had tightened substantially. A number of political-risk events, and the prospects of further to come, hardly dented the rally.
Emerging market asset classes were in especially high demand in the period under review, building on the momentum of previous months. Post-Brexit capital flows to emerging market economies (EMEs) strengthened further, boosting their equity prices and currencies and driving down their sovereign and credit spreads, which in some cases flirted with troughs seen in the third quarter of 2014. As the BIS international financial statistics indicate, EMEs' securities issuance had already rebounded in the first half of the year, despite more sluggish banking flows. Thus, the subsequent pickup in dollar issuance may have been sufficient to stop or even possibly reverse the first-quarter (slight) reduction in the stock of total dollar credit to non-bank borrowers in EMEs - a stock that had roughly doubled since 2009. This key measure of global liquidity conditions had declined to $3.2 trillion at the end of the first quarter from its $3.3 trillion peak in the third quarter of 2015 - even as the dollar credit to all non-banks outside the United States had picked up slightly, to $9.8 trillion.
Against this backdrop of generalised ebullience, bond markets seemingly pointed to economic weakness ahead - the first dissonant note. Bond yields continued to decline and yield curves flattened - typically a tell-tale sign of a low growth outlook. Adding to the sense of gloom was talk of the prospect that the global economy would be stuck in low gear into the distant future. And similarly discouraging were discussions of the need for permanently lower interest rates to counter secular weakness. While it has come down slightly recently, the overall stock of sovereign debt trading at negative yields reached new peaks in the summer, well above $10 trillion. Even more extraordinary, negative yields began to spread well into secondary corporate bond markets and even to reach some primary issues. And rates were also historically low relative to growth - a standard benchmark to assess their level.
At the same time, banks appeared to struggle - the second dissonant note. Earnings announcements ranged from mixed to disappointing, bank equity prices underperformed the overall index or fell, and price-to-book ratios generally languished below 1. Euro area and Japanese banks' valuations came under particular pressure, hovering at around 0.4, somewhat lower than those of UK banks and considerably below those of their US peers, which exceeded 1. The release of the euro area stress test results in July provided little relief, especially in jurisdictions with stubbornly high non-performing loans: markets hardly budged. Moreover, since July signs of tensions have appeared in the dollar short-term funding market: the key Libor rate has risen, partly in response to forthcoming regulation of US money market mutual funds that has raised the cost of uncollateralised funding. And, as analysed in detail in the Review, the cost of dollar borrowing through the vital FX swap market has risen even more for several currencies, notably the yen and the euro, tightening funding conditions for the banks that rely heavily on it.
There is, of course, a familiar explanation for this dissonance: the interaction of central bank decisions with market participants' portfolio choices. Through a combination of near zero, or even negative, policy rates, large-scale asset purchases and forward guidance, central banks have been seeking to ease financial conditions in order to boost the economy and, above all, to bring inflation up closer to their numerical objectives. The period under review was no exception, with further easing by the Bank of Japan, the Bank of England and other central banks, and signs of continued accommodation by the Federal Reserve and the ECB. For their part, market participants have depressed interest rates further, either in an attempt to seek out duration to hedge long-term liabilities (eg by pension funds and insurance companies) or to pick up vanishing additional returns. The result has been persistent ultra-low interest rates and even negative bond yields way along the maturity spectrum.
In turn, the prospect of lower rates for longer has had a dual effect. On the one hand, it has fuelled a familiar shift into equities and a broader search for yield, leading to the usual signs of exuberance. On the other hand, it has raised serious concerns about banks' profitability, as ultra-low rates and flat yield curves tend to erode their net interest margins and to reduce the cost of carrying non-performing loans, in turn delaying the necessary clean-up of banks' balance sheets.
There has been a distinctly mixed feel to the recent rally - more stick than carrot, more push than pull, more frustration than joy. This explains the nagging question of whether market prices fully reflect the risks ahead. Doubts about valuations seem to have taken hold in recent days. Only time will tell.
Developments in the period under review have highlighted once more just how dependent on central banks markets have become. And their sensitivity to central banks' every decision and utterance emerged in full force once again on Friday last week, when markets plunged and bond yields backed up following statements nuancing the prospects of the future course of policy in key jurisdictions. It is becoming increasingly evident that central banks have been overburdened for far too long. As argued in detail in the latest BIS Annual Report, a more balanced policy mix is essential to bring the global economy into a more robust, balanced and sustainable expansion.
Given recent developments in financial markets, it is fitting that the special features in this issue all tackle one aspect or another of global financial flows and capital markets, ranging from bonds to currencies to banking.
In the last few days, there has been a stirring from the tranquillity in financial markets that reigned for much of the summer, where stock market indicators pointed to unabated risk appetite - exuberance, even - on the part of market participants. Prices were at all-time highs, while volatility was close to record lows.
But those signs of investor confidence appeared side by side with market anomalies which are normally associated with stress in financial markets.
One such anomaly is the breakdown of "covered interest parity" - one of the best established laws in international finance, which states that the interest rate implicit in the foreign exchange market should coincide with interest rates in the money market. This relationship started to break down during the Great Financial Crisis of 2007-09. Since mid-2014, the gap between these two measures of the funding cost for the US dollar has widened further. Market players who borrow dollars in FX markets by pledging yen or euros pay more to borrow dollars than they would on the money market.
The violation of covered interest parity raises three questions:
- Why has the gap opened up and widened in recent months?
- Why does the gap not disappear through textbook arbitrage, where someone borrows at the low interest rate and lends at the higher interest rate?
- Should we be concerned by this breakdown of a time-honoured textbook rule?
As to why the gap has opened up, the findings in the special feature by Borio, McCauley, McGuire and Sushko suggest that the divergence of monetary policy across the major advanced economies has played a key role. Banks, pension funds and life insurance companies from those economies with low or negative rates have sought to pick up yield by purchasing dollar assets. The search for yield has taken on the character of a "flight from zero" as these institutions have compressed US yields, while their attempts to hedge the resulting foreign exchange risk have pushed the basis wider.
Perhaps more puzzling is why this gap has not closed in the usual way through arbitrage. The persistence of the gap suggests that banks and other financial intermediaries do not have enough capital available to take on such transactions, or at least are putting such a high price on the use of their balance sheet to make the trade uneconomical at these spreads. Since hedge funds or other unregulated entities are also reliant on dealer banks to put on leveraged trades, the banking sector remains the focus of attention.
Currently, banks face challenges to their business models due to low interest rates, but many of them are well capitalised, with capital levels comfortably exceeding any regulatory requirements. The fact that the gap persists in spite of such well capitalised banks suggests that regulation cannot be the whole story as to why banks are not engaged in these types of trades. Instead, the deviations from covered interest parity suggest that banks are putting a high internal price on balance sheet capacity in their capital allocation decisions, perhaps due to a less sanguine view of their prospects in an environment of low interest rates as well as to a keener appreciation of the potential risks.
This leads us to the third question: should we be concerned? The special feature does not address this question, but some reflections are possible. The persistent deviation from covered interest parity may not be a concern for policymakers in itself, but policymakers should take note of it as an indicator of the workings of the banking sector. If banks put such a high price on balance sheet capacity when the financial environment is largely tranquil, what will happen when volatility picks up? If they react to resurgent volatility by reducing their intermediation activity, as happened during the 2007-09 crisis, the banking sector may become an amplifier of shocks rather than an absorber of shocks. For this reason, it would be important to keep a close eye on this formerly rather esoteric corner of the foreign exchange market.
Let me now turn to the other chapters in this issue of the Quarterly Review. We are using this occasion to expand our data offering in five key areas, and are publishing a guide. The five new data offerings are:
- On our banking statistics, we are posting more details on the counterparty breakdowns in the locational banking statistics (cross-border positions in up to 29 countries, on counterparties in up to 200 countries).
- Time series on credit-to-GDP gaps (43 countries, starting as early as 1961).
- Commercial property price indicators (25 series for 10 jurisdictions).
- Historical time series on consumer prices (60 countries, for an average of 55 years), though in some cases going back 100 years or more.
- Daily data on nominal effective exchange rates (61 countries).
I have already mentioned the special feature by Borio, McCauley, McGuire and Sushko on the breakdown of covered interest parity. Let me give a brief overview of the other special features in this issue of the Quarterly Review.
The article on foreign exchange market intervention in emerging market economies (EMEs) by Domanski, Kohlscheen and Moreno looks at how EME central banks have adapted to new realities. Currency interventions are sometimes portrayed as attempts to keep currency values depressed for competitive advantage, but the authors find that cushioning against financial shocks has become much more important as a motive for intervening in currency markets. The new focus also has implications for the tools used. Those tools have included not just spot purchases but also forwards, options and swaps. The evolving role of central banks reflects, in part, the surge in issuance by private EME borrowers (as described in the following piece by Serena and Moreno), creating obligations and exposures that central banks will need to address when capital flows reverse.
The feature on offshore bond markets by Serena and Moreno looks at offshore bond issuance by EMEs. The authors' findings shine a light on the motives for offshore issuance. Emerging market firms can surmount the obstacle of underdeveloped domestic markets by issuing offshore, and global liquidity conditions in the past few years have allowed many firms to tap bond markets at low cost. But there is also a dark side to this apparent triumph. The proceeds of the bond issuance are parked in financial assets at home - possibly a form of carry trade - which could raise issues about financial stability if those positions had to be unwound under stress.
The special feature on euro-denominated cross-border bank lending by Avdjiev, Subelyte and Takáts examines the impact of the asset purchase programme of the ECB on cross-border lending in euros. The authors find that euro lending increased between country pairs where a higher share of cross-border lending was already denominated in euros - whether or not the countries were part of the euro area. They uncover the tell-tale pattern of an international currency where a currency depreciation is associated with greater cross-border lending in that currency. The euro has increasingly taken on the attributes of an international funding currency, just as the dollar assumed that role many decades ago.