QE inside out

Twitter can often be something of an intellectual cesspool in which people continue to shout across the cultural divide without ever hearing the views of the other side. It can, however, be a source of great enlightenment and I was very taken by a post (here) by Alfonso Peccatiello explaining why QE can never be inflationary. Having thought about it, there are many elements of this Twitter thread which are wrong. Nonetheless, it served a very useful purpose in forcing me to assess the nature of money and for that it is to be commended.

Inside versus outside money

At issue is the nature of “inside” versus “outside” money. Peccatiello argues that “inside money never reaches the real economy. Outside money does” and to the extent that he characterises QE as inside money, it cannot therefore impact on inflation. I suspect there is some confusion about definitions here and in order to get a better handle on this, we need to understand the concepts under discussion. The notion of inside versus outside money was introduced into economics by the seminal work of John Gurley and Edward Shaw, Money in a Theory of Finance, published in 1960. It was a very innovative work for its time and looked at the interaction between the real economy and monetary growth. As they put it, “real or “goods” aspects of development have been the center of attention in economic literature to the comparative neglect of financial aspects.”

Gurley and Shaw (G&S) define inside money as originating from within the private sector. Since one private agent’s liability is simultaneously another agent’s asset, inside money is in zero net supply within the private sector. Thus, funds held in bank accounts would be classed as inside money because they are an asset of private firms or households but a liability of private sector banks. Conversely, outside money derives from outside the private sector and is either fiat (unbacked) or backed by some asset that is not in zero net supply within the private sector. An increase in the stock of currency by the central bank would be classified as outside money, because although it is an asset of the private sector it is a liability of the public central bank.

The vexed question of money neutrality

Since outside and inside money represent different types of liquidity, they have differing effects on the wider economy. One implication of this is that the money stock is not homogenous – money is not just money as it is comprised of these two differing types. This calls into question the notion that money is neutral (i.e. it does not impact on real quantities in the long run and serves only to impact on prices). In light of the debate regarding the impact of liquidity creation by central banks and the recent surge in inflation, this may help us to shed some light on the monetary transmission process and the role of QE within it.

Without boring the reader with the details of the process, G&S demonstrate that in their general equilibrium framework inside money is neutral but outside money is not (the interested reader is referred here). How do we categorise QE?  In the sense that the central bank creates reserves to buy assets from the private sector (in this case, the non-bank private sector) but uses them to buy government bonds, they are merely swapping one type of outside money for another (cash for bonds). Consequently, the inside-outside composition of private sector assets remains unchanged. I thus agree with Peccatiello that QE can be categorised as inside money. If we accept the proposition of money neutrality, this implies that QE is unlikely to have any long-term impact on real activity. But this does not mean that it has no impact on prices. Indeed, classical monetary theory argues that if it does anything, QE will impact on price inflation.

Moreover, monetary neutrality relies on the absence of frictions but, as the work of Karl Brunner and Allan Meltzer has demonstrated, such frictions do exist and thereby allow monetary policy to have an impact on the real economy – perhaps only for a limited period of time. QE may not have had the impact on the economy that central banks hoped but there is in theory scope for it to boost activity. Analysis conducted by the BoE in 2011 suggested that the initial round of asset purchases boosted UK GDP by 1.5% to 2% compared to what would have happened in its absence.

Should central banks continue with asset purchases?

If we therefore accept that asset purchases do impact on inflation and prices, does it make sense for central banks to continue the programme, particularly in view of the surge in US inflation to 5.4% last month – the fastest rate since 2008? The short answer would appear to be no. Whilst it is true that the prices of used cars contributed a third of the monthly rise in the CPI in June, which is likely to be a temporary phenomenon, there is evidence that prices are rising rapidly across the board. Accordingly the Fed is widely expected to taper its asset purchases before too long with suggestions that the Fed will broach the subject at next month’s Jackson Hole symposium. Investors will recall the 2013 experience when the prospect of a slowdown in Fed asset purchases prompted the infamous taper tantrum which resulted in a spike in bond yields. However, with financial asset prices at red hot levels, taking some air out of the market may actually not be such a bad thing.

Last word

Whilst there is general agreement that QE has been the primary driver of asset prices in recent years, there remains much debate about its impact on the wider economy. Whilst I have never been persuaded of some of the claims made for it by policymakers, I have never accepted that it has zero impact. Even if the effects are small, such has been the magnitude of asset purchases that even small spillovers will show up in GDP and inflation data. This 2016 paper by Martin Weale and Tomasz Wieladek offers evidence that in contrast to the claim made by Peccatiello, “US (UK) QE had a similar (much larger) effect on inflation and (than) GDP.” If asset purchases have helped the economy to avoid a slow post-pandemic recovery, they have done their job. But having done their job, it may now be time to think about scaling back.

Telling it like it is

Andy Haldane, the Bank of England’s outgoing chief economist, is an eclectic thinker on matters of economic policy and his valedictory speech on his last day in the role was a tour de force of the issues facing central banks today. In recent months, Haldane has warned that inflationary pressures are building and in each of the last two MPC meetings he has voted to reduce the BoE’s asset purchase limit from the planned £895 billion to £845 billion (though I sometimes wonder whether an impending departure allows policymakers to throw off the shackles and vote against the consensus). Whilst this is not a huge change in the grand scheme of things it is a signal of intent and in his speech Haldane explained the reasoning behind his thinking.

In Haldane’s view the current conjuncture is rather different to that which prevailed in the wake of the GFC when the recovery was a rather slow and protracted affair. This supported the slow withdrawal of the post-2008 stimulus but with the economy apparently set to grow very strongly in 2021 “this time that policy script feels stretched.” The danger is that large and rapid balance sheet expansions (chart) but limited and slow withdrawals “puts a ratchet into central bank balance sheets” which raises concerns of fiscal dominance – the extent to which monetary policy becomes subservient to the needs of the fiscal authorities to manage deficits and debt rather than the primary goal of fighting inflation. Economists agree that large volumes of asset purchases do increase the risk of fiscal dominance but are less in agreement that recent actions will lead to such an outcome. There have been suggestions, particularly in Germany, that the ECB’s policies in recent years have been designed to support the heavily-indebted euro zone economies although we will only know if fiscal dominance has taken root if the ECB fails to act in the face of any inflation pressure. But as Haldane put it, “this is the most dangerous moment inflation-targeting has so far faced.”

He also pointed out that “a dependency culture around cheap money has emerged over the past decade.” I have a lot of sympathy with this view and argued strongly that central banks should have been quicker off the mark from 2013 onwards to withdraw some of the monetary stimulus put in place after the GFC. This is not to say that policy should have turned restrictive but it perhaps should have been less expansionary. After all, as I have consistently pointed out, the economy in 2012/13 may not have been in great shape but it was in far better condition than in 2008/09. The actions of central banks in the five years after the GFC were akin to performing life-saving surgery on a patient long after they had left the operating theatre and were resting on the ward.

As with all procedures there are side effects and one of those associated with lax monetary policy is extremely high asset values, both financial (equities) and real (housing). If investors do not expect central banks to take some air out of the balloon, they will continue inflating prices. This is more than just a financial market problem – there are social implications as well. For example, many young people who were leaving school in 2008 will now be thinking of starting families but find themselves priced out of the housing market. Moreover, rising asset values are fine for those sitting on a portfolio of stocks but serves to enhance the wealth disparity versus those who do not. And as I have been pointing out for some time, low interest rates penalise savers – particularly those nearing retirement. Expansionary monetary policy is undoubtedly the right policy in the right conditions but I am with Andy Haldane in hoping that central banks do not wait too long before cutting back on some of the support they are currently providing.

Haldane’s speech stood in contrast to the BoE Governor’s annual Mansion House Speech, given a day later, which was a rather more conservative take on the state of the economy. That said Andrew Bailey did highlight the upside risks to inflation, and I agree with him that they are likely “to be a temporary feature of the bounce-back.” Bailey’s speech carried the usual warning from policymakers that the central bank is prepared to tighten monetary policy if required to stave off an inflation threat but that this does not form part of the current baseline. Whilst inflation has not proven to be a problem in the wake of the GFC, so that this threat has never been put to the test, there may come a point where the BoE may have to follow up on its threat to take away the punchbowl.

There are some former BoE policymakers (notably Danny Blanchflower) who are completely opposed to the idea of raising interest rates any time soon, and who give no credence to the Haldane view of the world. But this is to assume that the conditions prevailing today are similar to those more than a decade ago. But this is not the case: As Bailey pointed out in his speech the degree of economic scarring following the recent output collapse has been far smaller than expected. Moreover, governments are adopting a much more active fiscal policy approach today, in contrast to a decade ago when monetary policy had to do all the heavy lifting.

A final argument as to why the zero (or negative) interest rate policy should have a limited shelf life is derived from the Japanese experience where more than 20 years of expansionary policy have done little to get the economy back on the path which the government desires. There is a general belief that monetary policy is neutral in the long-run i.e. it may impact on nominal quantities but does little to influence real growth rates or productivity. This has been borne out by the Japanese experience which demonstrates that monetary policy does nothing to impact on the supply side of the economy. This is hardly surprising: Monetary policy cannot boost the capital stock or improve education and training standards, implying that other policy prescriptions are required. The UK’s dire productivity performance over the past decade makes it clear just how much these alternatives are sorely needed.

As Andy Haldane goes off to face new challenges as Chief Executive of the Royal Society of Arts he will leave a gap at the BoE which will not easily be filled. He has not met with universal acclaim from within the economics profession but I have always found his willingness to cross-pollinate economic ideas with those from other disciplines has resulted in some illuminating insights. He will leave big shoes to fill.

Rethinking low interest rates

Over the years, prevailing economic orthodoxy has tended to follow fashions with policy makers pursuing a particular course of action only to subsequently switch tack and repudiate what has gone before. Very few people today believe that fixed exchange rates are a good idea (unless you happen to be in the euro zone where the rates of member countries have been fixed against each other for more than 20 years). Quaint ideas such as targeting money supply have also fallen from favour. Even the notion of using fiscal policy as a countercyclical tool, which was banished from the lexicon in the 1980s (apart from a brief reappraisal in 2008-09), is now part of the policy armoury. I thus wonder how long it will be before central bankers revisit the question of whether low interest rates do as much harm as good.

It has long been my view that central banks around the world made a policy error in not normalising monetary policy more swiftly in the wake of the 2008 crash. Although the contraction of global liquidity in the wake of the Lehman’s bankruptcy warranted a massive monetary response, there were few good reasons to justify why monetary conditions in 2012 were required to be quite as easy as those prevailing at a time of financial Armageddon. The Riksbank was one of the few brave enough to begin a tightening cycle in 2010 but the Swedish economy was caught in the backwash of euro zone turbulence and the central bank was forced to cut rates even below post-Lehman’s levels. I have often suspected that this policy about-turn deterred other central banks from making similar moves ahead of the Fed or ECB.

This is not to say that global interest rates needed to go back to pre-2008 levels. Factors such as demographics and the sharp slowdown in productivity growth justify a lower equilibrium real rate. However, one of the things economists warned about was that holding rates too low reduced the scope for conventional monetary easing in the event of an exogenous shock. That shock duly arrived in the form of the Covid pandemic and given the limited scope for rate cuts, central banks were forced to swell their balance sheets to unprecedented levels. This has opened up a whole can of worms. On the one hand it has sparked inflation fears whilst on the other it has led to significant market distortions, pushing up both bond and equity markets.

The inflation issue

Dealing first with inflation, this is a subject which has been at the top of the agenda throughout this year with the US 10-year yield hitting a 12-month high of 1.74% in March versus 0.91% at the end of 2020. At his annual shareholders meeting at the weekend, Warren Buffett warned that “we are seeing very substantial inflation.” Joe Biden’s huge US fiscal expansion plans have further raised concerns that the economy may overheat and Treasury Secretary Janet Yellen’s suggestion that “it may be interest rates will have to rise somewhat to make sure that our economy doesn’t overheat” sent ripples through equity markets which have been driven to record highs on the back of ultra-expansionary monetary policy.

However, we may be overestimating the link between monetary policy and inflation. The academic literature is unambiguous that there has been a change in the inflation process over the past 20 years. There is less agreement on whether that represents a weakening of the link between inflation and activity growth or whether the decline in inflation volatility is due a reduction in the volatility of economic shocks. In my view the former explanation counts for more in a world in which the rise of China as a major production centre has changed the dynamics of the global economy. That being the case, central bank actions in the industrialised world have played less of a role in driving down inflation than they like to believe. Accordingly, they may have less power to prevent any significant acceleration. Furthermore, if the link between inflation and the economy is less well defined than many suppose, it is harder to justify low interest rates on the basis of low inflation.

Are prolonged ultra-low rates even effective?

Whether inflation does or does not accelerate is not the focus of debate here. The bigger concern is that central banks appear fixated only on inflation as a measure of determining whether interest rates are at appropriate levels whilst ignoring a number of other factors. Indeed whilst central banks are right to take their inflation mandate seriously, there are a number of downsides associated with an ultra-easy monetary policy.

Starting with the bigger question, how sensitive are industrialised economies to interest rates anyway? My own modelling work always came to the conclusion that the UK was never particularly sensitive to interest rate moves. More detailed academic work by Claudio Borio and Boris Hofmann at the BIS suggested that monetary policy tends to be less effective in periods of ultra-low interest rates. They noted further that “there is also evidence that lower rates have a diminishing impact on consumption and the supply of credit.” Two reasons were given for this: (i) the conditions which prompted a cut in rates to the lower bound in the first place (a balance sheet recession) generate economic headwinds which make recovery more difficult and (ii) the impact of low rates on banks’ profits and credit supply generate feedback effects which impede recovery. We only need ponder the Japanese experience of the past 20 years to realise there may be something in this.

The market impact

Although the economy in the industrialised world has not boomed in the last decade, equity markets clearly have with the Shiller 10-year trailing P/E measure on the S&P500 last month trading at a 20-year high of 36.6x. Prior to the March 2020 collapse I did note that the prevailing level of 31x pointed to a market that was too expensive but we are now at the second highest levels in history (chart below) – lower than in 1998-2001 but above 1929 levels.

This in turn raises a question whether central banks have a duty to take account of financial asset prices in their monetary policy deliberations. Former Fed Chairman Greenspan used to take the view that we could not spot a market bubble until it had burst and that the role of monetary policy was to mop up after the fact. That view no longer holds since we can clearly identify that US markets are in bubble territory. To the extent that central banks are increasingly responsible for financial stability it is incumbent upon them to ensure that the banks which they monitor will not be adversely impacted by a market correction. In fairness, banks are subject to regular and rigorous stress tests and central banks are confident that capital buffers are sufficiently large to withstand a major market shock.

However, the gains from high asset valuations generally accrue to high income households which has distributional consequences for the economy. Central banks can rightly argue that this is not part of their mandate and is therefore not something they have to worry about. But to the extent that governments, which set the mandate, do care about distribution there is a case for central banks to at least think about this problem before it is forced upon them. Then of course there is the ongoing problem of what low interest rates do to savers, particularly those with an eye on retirement – a problem I have highlighted on numerous occasions. Most people do the bulk of their retirement saving in the last ten years before they leave the workplace – precisely the time when they are advised to reduce equity holdings and overweight their pension portfolio towards fixed income. Pension annuity rates remain nailed to the floor in this low interest rate environment (chart below) which means that retirees get a much smaller payout for any given pension pot than they did in the past. Low rates clearly have generational consequences.

Last word

I am not advocating that central banks should imminently raise interest rates. Indeed any such move is only likely to occur well into the future. But in a post-Covid environment where fiscal policy is likely to be much looser than in the wake of the GFC, there will be less need for monetary policy to do most of the heavy lifting. This should at least stimulate a proper debate about the pros and cons of ultra-low (and in some cases negative) interest rates which has been lacking over the past decade.

A new monetary paradigm

Last week marked a big event in central banking history and economists will look back at Jay Powell’s speech on 27 August as the point at which the parameters of the three decade experiment with inflation targeting were changed. Central banks have tended to adopt a numerical target for inflation, usually centred around 2%, but in recent years inflation has undershot this target. A variety of reasons have been put forward for this. Many central bankers will argue that the adoption of inflation targeting was one of the factors which squeezed the high inflation of the 1970s and 1980s out of the economy. However, we can also add a number of exogenous factors such as the opening up of China as the workshop of the world, the end of the Cold War and more flexible labour markets.

Whatever the reason, there is a strong case for suggesting that the current regime which emerged in the wake of the high inflation of the 1970s and 1980s is no longer appropriate. For one thing, as currently practiced, inflation targeting contains an inbuilt asymmetry since central banks are more likely to react to inflation overshoots rather than undershoots for fear of being seen as soft on inflation, despite the fact that the target usually allows for some leeway around the central case. But we find ourselves today in an environment in which economies have taken the biggest hit in at least 90 years and inflation concerns are rapidly being put on the back burner as policy makers start to concern themselves with issues such as unemployment. Accordingly, the Fed announced last week that it will in future target maximum employment and “following periods when inflation has been running persistently below 2 percent, appropriate monetary policy will likely aim to achieve inflation moderately above 2 percent for some time.” In effect, it will adopt a medium-term price level target – something many economists have devoted time to thinking about over the past decade.

This has a number of important implications for monetary policy in future. For one thing it effectively means that the Fed will be less likely to pre-emptively tighten policy in response to a perceived inflation pickup after a period of below-target inflation. At a time when real economy concerns are paramount this makes a lot of sense. But questions arise as to what the Fed means by the two parts of the phrase “inflation moderately above 2 percent” and “for some time?” Much of the early discussion about price targets tended to focus on short-term deviations of inflation from the target. Thus if an exogenous shock results in inflation averaging 1% over a year, inflation can run at 3% the following year in order that it averages 2% over the whole period. But if inflation averages 1% over a 10-year period, this would require a 3% rate over the next ten years in order to get the long-run average inflation rate back to 2% (chart). In such a case is 3% “moderately above 2%?” And does a ten year period represent “some time” or does this represent a whole different paradigm?

As I argued in this post once we start to tolerate higher inflation the concept of price stability begins to become eroded. There is thus a concern that the price level targeting regime could result in inflation expectations becoming de-anchored. We might not worry about this any time soon but it is a discussion which will doubtless be taking place within the Fed. That said, at a time when government debt levels have gone through the roof in many countries, no government is going to complain about a little bit more inflation if it helps to alleviate the debt burden.

One of the less discussed aspects of the change was that on the rules versus discretion spectrum, which has been a feature of monetary policy over recent decades, the Fed’s policy announcement is very much a step in the direction of discretion. It ascribes a greater weight to “the shortfalls of employment from its maximum level” where the Fed freely admits that “the maximum level of employment is a broad-based and inclusive goal that is not directly measurable and changes over time.” Full employment has always been a nebulous target, but it is about to become even more so and runs the risk of being whatever the Fed believes it to be in order to satisfy its current policy stance. For markets which have become used to interpreting mechanistic rules, this implies a rather more opaque central bank than we have become used to. It also suggests an enhanced role for forward guidance which is likely to become even more important as a policy tool.

Another important aspect of the new policy is that the central bank will be even less concerned in future with trying to fine tune the economy via monetary policy, since shortfalls imply a persistent deviation from the nebulous concept of maximum employment whereas deviations imply cyclical divergences. Whilst the Fed’s moves make a lot of sense from an inflation perspective, it also fits with the new realities of the post-Covid economy in which central banks will have to play their part in dealing with the resulting economic scarring. By explicitly committing to keeping rates low(er) for longer, the Fed is perhaps taking the first steps along the path towards financial repression. 

However, it is important to be aware of some of the downsides. For one thing a policy of targeting price levels implies the potential for short-term inflation volatility if economic agents are not sufficiently forward looking and base their inflation expectations on extrapolation of current rates out into the future rather than the longer term goal of ensuring that price levels will eventually return to target. In addition, the standard objection in the literature is that there is likely to be greater output volatility under a price level targeting regime as the central bank does not attempt to moderate cyclical swings as frequently.

Nonetheless, it is likely that other central banks will eventually follow in the Fed’s footsteps and adopt a version of an average price level target with a focus on the real economy. However, it is unlikely to find much favour in the euro zone where the focus on short-term price stability remains paramount. But as Martin Sandbu points out in the FT the ECB has “treated its own legal mandate far too narrowly … There is a widespread misperception that the ECB is treaty-bound to the single duty of ensuring price stability … But beyond this, the ECB has a legal obligation to ‘support the [EU’s] general economic policies’” (a point I have made repeatedly in the case of the BoE). The ECB may not be a slave to monetary policy fashions but it is not the Bundesbank and has proved itself pretty flexible in its monetary dealings in the past. It just might take a while to get there.

Stay positive or turn negative?

In recent weeks the Bank of England has given the impression that it may be prepared to take interest rates into negative territory. Although the debate has gone a bit quiet of late, it has not gone away, perhaps because the BoE believes the economic collapse in April was not quite as bad as previously expected or, and this is my preferred take, it was only ever a device to jawbone market interest rates as low as possible.

The economist Silvio Gesell was one of the first proponents of negative interest rates in the nineteenth century when he proposed a tax to dissuade people from hoarding cash. But it was never seriously tried in a policy context until the Swiss introduced a policy of negative rates in the 1970s in a bid to prevent foreign investment flows from driving the franc higher. However the policy was deemed a failure and the idea was eventually abandoned in 1978 after an experiment lasting six years. In the wake of the 2008 financial crisis the idea came back onto the agenda with Denmark being the first mover, again as a means to hold the currency stable. This time, other countries followed suit with the likes of Switzerland (again), the euro zone, Sweden and Japan all driving rates into negative territory. Both the US and UK have resisted the charms of negative rates, largely because there is a presumption in the Anglo Saxon world that the monetary transmission mechanism ceases to operate properly with interest rates below the lower bound.

There are many good arguments against negative rates

In the current policy framework, the banking sector is charged a negative interest rate on its cash deposits at the central bank. Banks have an incentive to run down their cash holdings and either lend more into the wider economy or pass on the negative cost to their customers who run down their money holdings, and in the process stimulate the economy as they spend their cash. However, low interest rates and flat yield curves distort time preferences for households and companies, which results in sub-optimal resource allocation (e.g. they allow zombie companies to operate which would otherwise cease trading). Central bankers who have observed the experience of Japan over the last two decades cannot be blamed for calling into question the usefulness of ultra-lax monetary policy. Ironically, in the late-1990s I remember half-jokingly suggesting to a Japanese economist that the BoJ should consider negative rates. It was probably not my greatest idea in retrospect.

Furthermore, ultra-loose monetary policy distorts markets. By reducing the returns to cash holdings, investors have an incentive to seek higher returns by loading up on risky assets, which in turn results in widening disparities between market prices and fundamentally justified levels. No investor would question the view that low rates have helped markets to blow out.

For all that, there is no good reason in theory why interest rates should not go negative.  After all ancient mathematicians regarded negative numbers as “false” and it was not until the late seventeenth century that respectable mathematicians such as Leibniz began to take them seriously in Europe. Today, however, we are all familiar with the concept and we might wonder why they were ever regarded with such suspicion.  Moreover to the extent that real economic quantities respond to real interest rates, we have long become used to the notion of negative real rates with nominal interest rates lower than inflation.

Yet there is something of the taboo about a negative nominal interest rate. Perhaps one reason is that the interest rate represents the cost of time: it represents the return derived from waiting; from saving rather than consuming. Perhaps it offends the Puritan streak in the western psyche. Or maybe because the arrow of time only runs in one direction, a negative interest rate somehow inverts the cost of time and is therefore perceived as unnatural. Whatever the reason, many people have difficulties with the concept of negative central bank rates.

For all the evidence amassed by the likes of the ECB suggesting that negative rates have helped to stimulate the economy, we should treat the arguments with a pinch of salt. Without any doubt, negative interest rates penalise savers. Unless we are forced to work long past our planned retirement date, we all need to make provision for old age and this is made all the harder by low or negative interest rates. There may be an argument in favour of temporarily trying to boost the economy by cutting rates into negative territory but the ECB has held the depo rate below zero for six years. I fear that a prolonged period of negative rates will ultimately prove counterproductive as individuals attempt to raise their precautionary saving.

But consider this …

One of the key features of all the countries that have experimented with negative rates so far is that they run a current account surplus i.e. there is a surplus of domestic saving with respect to investment (chart). Both the UK and US run current account deficits – they suffer from deficient domestic saving. At first glance, you may ask whether negative rates in these economies are such a good idea if they encourage further dissaving. In a static framework, they are not. But let’s try to think through the dynamics. Encouraging households to bring forward spending should widen the current account deficit in the near-term and ought, in theory, to result in currency depreciation. This in turn should generate higher imported inflation, which after all is how central banks have justified their actions, and allow them to respond by returning interest rates towards positive territory.

In this framework the key transmission mechanism is the exchange rate. If the cut in interest rates is not sufficient to produce concern in the FX market, the negative interest rate policy will not have the desired effect. This might be because the pickup in consumption is insufficient to generate a current account deficit so the currency market remains unconcerned. Or it might be due to the fact that the currency in question (the likes of the Swiss franc, yen and euro) acts as a safe haven, particularly since rates elsewhere are also extremely low. In either of these cases, perhaps interest rates will have to be pushed so far into negative territory to have the desired effect that the side effects would be unacceptable. But this begs the question whether they would work better in an economy which already runs a current account deficit, and where the FX market is perhaps more sensitive to external deficit concerns. The pound would certainly be such a candidate.

I would be hesitant to advocate the BoE cutting interest rates into negative territory because the experience elsewhere shows that once they go below the zero line, it proves difficult to get them back up again. But if I were a maverick on the MPC I would at least try to ensure that this argument gets a hearing and make the case for a short, sharp dip into negative territory with the unspoken assumption that they will be raised after (say) two years. There is nothing to be gained by holding rates below zero for long. But there also seems little to be gained from a prolonged period of holding them so close to zero they might as well be negative.