Andrew Haldane: Halfway up the stairs
The views expressed in this speech are those of the speaker and not the view of the BIS.
"Halfway up the stairs
Isn't up,
And it isn't down.
It isn't in the nursery,
It isn't in the town."
AA Milne, Halfway Down (1924)
It is the 25th anniversary of Central Banking. So it is a good time to be asking some big questions of central banking. How have central banks evolved over the last quarter of a century? How has the crisis affected that evolution? And what lies in prospect for them over the next 25 years?
AA Milne's 1924 poem, Halfway Down, is a fitting description of the position central banks find themselves in today. During the past 25 years or more, central banks' mandates and instruments have moved upwards in steps. They have ascended the stairs.
But where this leaves central banks today is not entirely comfortable. Halfway up the stairs is neither up nor down, neither nursery nor town. That begs a natural question about "where next" for central banks over the next quarter-century.
Ascending the stairs
History will show the 20th century to have been an unprecedented boom period for central banking. At the start of that century, central banks were relatively few in number. For those that did exist, their responsibilities were heavily constrained. Often, their role was to act as no more than an operational agent for the government of the day, managing money, debt and foreign exchange markets. They were agents, not principals.
Through the course of the 20th century, central banks' influence began to grow. So too did their numbers. For some, this evolution began with being granted monopoly rights over issuing legal tender - fiat money. This gave central banks special status relative to commercial banks. It underscored their stability role as liquidity provider to the banking system: stabilising short-term money markets in normal times, acting as lender of last resort in abnormal ones.
It also secured, by fiat, a demand for central bank liabilities. This meant central bank money became the natural settlement medium, and central banks the natural clearing house, when discharging obligations among banks. As the marginal supplier of liquidity, this gave central banks the ability to set the marginal price of liquidity - the short-term interest rate. In that way, monetary policy was born and an important central bank step had been taken.
The early phases of monetary control by central banks were often inauspicious, highly constrained, or both. Monetary regimes during the latter part of the 19th century and the first part of the 20th were long on rules and short on discretion. The Gold Standard, and later the Bretton Woods regime, tethered monetary policy to a commodity or dollar anchor, with little scope for monetary discretion. And where monetary discretion did operate, it was not always a success. The Federal Reserve's discretionary monetary actions during the Great Depression were a salutary lesson.
But the breakdown of the Bretton Woods system in the 1970s brought a further sea-change. Tight international rules gave way to national discretion. With the accompanying rise of "monetarism", these monetary frameworks were often based around targets for the money supply. In most cases, the role of central banks was still to act as operational agent. But the technical nature of these regimes gave central banks a say in their design. A second step had been taken.
The "great inflation" of the 1970s, together with repeated failures to hit money supply targets, began to change that picture further during the 1980s. Monetary discretion was found to have come with costs. In particular, it had given birth to what economists subsequently called the "time-consistency problem". This problem arose from the temptation to over-supply money or over-relax regulations today, often to finance wars or to win elections, at the expense of inflation tomorrow. This problem was thus endemic among elected officials.
The birth of independence
As arm's-length technocrats, without the need to finance wars or win elections, central banks were seen as a bulwark against these time-consistency concerns. Thus was born central bank independence. If decision-making power over monetary policy were delegated to central banks, not as operational agents but as principals, the time-consistency problem could be curbed at source. With this argument, central bank independence spread rapidly and globally during the 1980s and 1990s. And with it, central banks had taken another significant step.
At the same time as monetary policy discretion was being granted, constraints were being placed around it. These were a necessary part of democratic accountability. These constraints included government-set success criteria - for example, inflation targets. In the jargon, many monetary policy regimes became characterised by instrument-independence but target-dependence. They combined rules with discretion - hence "constrained discretion".
Yet as significant was the accompanying shift in central bank transparency. This was seen as a necessary concomitant of delegated power. A century ago, the prevailing central bank ethos allegedly came from the then-Bank of England (BoE) governor, Montagu Norman: "never apologise, never explain". Then, secrecy and mystique were believed to be key weapons in central banks' armoury. Media appearances were rare. Central bank actions were confined to the back pages of newspapers, if at all. At the BoE, the aim was to keep the Bank out of the press and the press out of the Bank.
With independence granted, all that changed. Politicians and the public demanded explanations - and sometimes apologies - for policy choices. Policy meetings were pre-announced and minuted. And regular publications, such as inflation reports and financial stability reports, set out the analyses underpinning these decisions. Media appearances and newspaper articles began to appear. Central banks moved from the back to the front pages of newspapers. Journalists were no longer locked out, but instead ushered in for press conferences. Another step up the stairs had been taken.
End of the "Great Moderation"
This brave new world looked set to continue indefinitely during the early years of this century. Inflation was low and output stable, the so-called "great moderation". The Holy Grail of growth without risk seemed to have been discovered. Yet moderation was sowing the seeds of its own destruction, dulling the risk senses of central banks, regulators and market participants. In this vacuum, financial risk-taking rose to Himalayan heights.
The global financial crisis of 2008 brought that decisively to an end. The financial system and economy tumbled from their Himalayan peaks. Great moderation gave way to "great recession", the sharpest and longest-lasting since the 1930s. Central banks quickly acknowledged their pre-crisis blind spot - ballooning debt among both banks and non-banks. And, not for the first time in their history, they reinvented themselves.
On the monetary policy front, extraordinary times resulted in extraordinary monetary measures. Interest rates globally were lowered to levels previously never seen, accompanied in some major advanced countries by purchases of government securities - so-called quantitative easing (QE). In some countries, central banks in addition purchased private sector assets, such as mortgage-backed securities and corporate bonds - so-called "credit easing". Others, including the Federal Reserve and Bank of England, issued forward guidance to help clarify their future policy intentions and thereby shape the yield curve.
This was terra incognita for central banks. The focus of monetary policy, at least in the 25 years prior to the crisis, had been on controlling the short-term safe rate of interest. Crisis-induced monetary interventions widened that focus significantly. By pursuing QE and forward guidance, central banks began to exercise influence beyond overnight rates of interest. And for those central banks that engaged in credit easing, this influence extended further, from safe to risky rates of return across a broad spectrum of assets. These were giant steps.
A second aspect of the reinvention was macro-prudential regulation. Inflation-targeting was necessary but, by itself, insufficient to curb the financial cycle. The response of governments has been to grant central banks new powers, focused on the needs of the financial system as a whole and towards the needs of the nonfinancial economy, as much as the financial sector. That is what the "macro" in macro-prudential meant.
Macro-prudential policy is gaining ground every bit as quickly as central bank independence did in the 1990s. It has quite radical implications. Pre-crisis credit cycles were allowed to operate largely unconstrained. Macro-prudential policy overturns that orthodoxy, with policy instead leaning against the credit cycle to moderate its fluctuations, both during the upswing and the downswing. It, too, is a big step forward.
A third aspect of the reinvention is in the operations of central banks. During the crisis, central banks flexed their balance sheets as never before. New facilities were introduced that extended liquidity for longer durations against expanded sets of collateral (public and private) to new counterparties (bank and non-bank). This took last-resort lending to a new level. Some central banks went one step further, becoming effective market-makers of last resort in some assets to secure market liquidity. These were new and bold steps.
A fourth dimension was transparency. With monetary, regulatory and operational policies all working in overdrive, central banks have had plenty of explaining to do. During the crisis, their actions have shaped the behaviour of pretty much every financial market and institution on the planet. So central banks' words resonate as never previously. Rarely a day passes without a forensic media and market dissection of some central bank comment. These, too, are significant steps to have climbed.
Where does this leave central banks today? We are not in Kansas any more. On monetary policy, we have gone from setting short safe rates to shaping rates of return on longer-term and wider classes of assets. On regulation, central banks have gone from spectator to player, with some granted micro-prudential as well as macro-prudential regulatory responsibilities. On operational matters, central banks have gone from market-watcher to market-shaper and market-maker across a broad class of assets and counterparties. On transparency, we have gone from blushing introvert to blooming extrovert. In short, central banks are essentially unrecognisable from a quarter of a century ago.
Up or down?
The key question, then, is where next? Will the next 25 years see as large a metamorphosis, from sleepy caterpillar to winged butterfly, as the last? It is impossible to know for sure. But it is interesting to attempt a bit of futurology and sketch some hypothetical evolutionary central bank paths. For the sake of illustration, let me consider two - one descending the stairs from whence central banks came, another ascending them to a whole new level.
One possible future path is that central banks simply retrace their steps, going back to basics. Extra-ordinary times - from great moderation to great recession - necessitated extra-ordinary measures. But extra-ordinary should not become the new ordinary. What might the new ordinary look like?
For monetary policy, it would probably mean a return to plain-vanilla inflation-targeting using conventional short-term interest rate tools - the one instrument, one target world. Direct use of central banks' balance sheets, to affect quantitative or credit easing, would return to being an emergency room measure, for use only when the zero lower-bound on interest rates bit tightly. Active monetary policy would mean small, infrequent touches on the tiller, perhaps accompanied by small, infrequent explanations. Former BoE governor Mervyn King aspired to make monetary policy "boring". In this scenario, his wish would be granted.
Regulation would follow a similar path. While not returning to their pre-crisis norms, regulatory rules would move to a more settled pattern. The very definition of a resilient financial system is one that does not require active intervention. With greater amounts of capital and liquidity in the financial system, this may be more realistic. There may be less need to fine-tune the regulatory engine and correspondingly less need for the tug-of-war between regulator and regulated that has characterised the past few years. A robust banking system ought also to be less prone to credit booms and busts, thereby reducing the need for discretionary macro-prudential actions to lean actively against the wind. Regulation would be handed back to the risk managers.
In this world, the operational activities of central banks could also return to a more even keel. Wider collateral, longer tenors and new counterparties were necessary ingredients during crisis. But a return to more conventional operations - short-term lending against government collateral to banking counterparties - may be sufficient to oil the wheels in normalised market conditions, as it was for hundreds of years prior to the crisis. Market-making, at least beyond short-term money markets, would be a last resort.
As for central bank transparency, with monetary, regulatory and operational policy no longer hyperactive, it is conceivable central banks could retreat from centre stage. Central banks could migrate to the back pages of newspapers. Like good children, they would be relatively rarely seen and infrequently heard. Central bank-watchers would find something better to do with their lives than hang on to every gubernatorial semi-colon.
For a central banker, this sounds like the good life - a low-intervention, low-visibility world. In some respects, it would be back to the future. The issue is whether this world, however desirable, is realistic. There are some reasons for doubt.
Financial system reinvention
One of the likely consequences of the crisis, and the resulting regulatory response, is that the financial system will reinvent itself. Financial activity will migrate outside the banking system. And with that move, risk may itself change shape and form. What previously had been credit and maturity mismatch risk on the balance sheet of the banking system may metastasize into market and illiquidity risk on the balance sheets of non-banks. This could have important implications for the stability of the financial system and the broader economy.
With more activity outside the banking system, and with the banking system itself better protected, the financial system and economy may become less prone to the low-frequency, high-cost banking crises seen in the past. But that is not the end of the story. Risk, like energy, tends to be conserved not dissipated, to change its composition but not its quantum. So it is possible the financial system may exhibit a new strain of systemic risk - a greater number of higher-frequency, higher-amplitude cyclical fluctuations in asset prices and financial activity, now originating on the balance sheets of mutual funds, insurance companies and pension funds. These cyclical fluctuations could in turn be transmitted to, and mirrored, in greater cyclical instabilities in the wider economy.
In this world, it would be very difficult for monetary, regulatory and operational policy to beat an orderly retreat. It is likely that regulatory policy would need to be in a constant state of alert for risks emerging in the financial shadows, which could trip up regulators and the financial system. In other words, regulatory fine-tuning could become the rule, not the exception.
In this world, macro-prudential policy to lean against the financial cycle could become more, not less, important over time. With more risk residing on non-bank balance sheets that are marked-to-market, it is possible that cycles in financial assets would be amplified, not dampened, relative to the old world. Their transmission to the wider economy may also be more potent and frequent. The demands on macro-prudential policy, to stabilise these financial fluctuations and hence the macro-economy, could thereby grow.
In this world, central banks' operational policies would be likely to remain expansive. Non-bank counterparties would grow in importance, not shrink. So too, potentially, would more exotic forms of collateral taken in central banks' operations. Market-making, in a wider class of financial instruments, could become a more standard part of the central bank toolkit, to mitigate the effects of temporary market illiquidity droughts in the non-bank sector.
In this world, central banks' words and actions would be unlikely to diminish in importance. Their role in shaping the fortunes of financial markets and financial firms more likely would rise. Central banks' every word would remain forensically scrutinised. And there would be an accompanying demand for ever-greater amounts of central bank transparency. Central banks would rarely be far from the front pages.
Which path?
Which central bank path is the most likely over the next 25 years? Your guess is almost certainly as good as mine. AA Milne ends his 1924 poem:
"And all sorts of funny thoughts
Run around my head,
It isn't really anywhere
It's somewhere else instead!"
A quarter of a century on, I am confident central banks will be "somewhere else instead". I am much less confident precisely where, whether upstairs or down. Central banking may need to await the 50th anniversary of Central Banking to find out.