I decided to launch my own blog under a non de plume I have been using for over 15 years - first at Bankers Trust - and then as an occasional and strictly amateur wine writer. Comte Flaneur is a fictional French aristocrat and playboy who enjoys the finer things in life.
I decided to publish this article I wrote a month ago, which I recently tweaked. This is a much longer post that would be normal, where I will comment on economic issues, current affairs, wine and anything else I care to vent my feelings on.
Preparing to launch QE2
A dangerous and reckless gamble
October 2010
As the Fed prepares to launch ‘QE2’ there is pervasive skepticism as to whether it will work, while some argue that the risks and unintended adverse consequences may outweigh any fleeting benefits. While one has to have serious reservations about QE in particular and the Fed’s policy reaction function in general, the chances of QE2 being seen to work – reflected in a period of above trend real GDP and significant nominal GDP acceleration over the next year are much higher than is generally believed. This may well be because the economy is picking up anyway after a temporary soft-patch and it has nothing to do with QE2; but Bernanke will still look like a genius. On the other hand there are plenty of things that could go wrong especially in negotiating the exit, in which case he will be pilloried and soon become a pariah. Most likely the Fed’s reckless experiment with money printing – because this is what it amounts to – will end badly. Moreover, the less the economy responds to new doses of steroids the greater the likelihood of an accident. QE is a desperate gamble by an institution guilty of myopic group think over a quarter of a century. In light of the disappointingly tepid recovery, stubbornly high unemployment and an underlying inflation rate deemed to be too low the Fed appears to be about to embark on another round of quantitative easing or ‘QE2.’ What is quantitative easing, or ‘QE’, and will it work? What are the possible unintended consequences and likely impact on markets of the implementation of QE? Is it equivalent to ‘printing money’ or a Ben Bernanke ‘helicopter drop’? Will the end game be hyperinflation or will central banks be powerless to prevent an slide into deflation? In this paper I will attempt to answer these questions with the caveat that we are sailing in largely unchartered waters, with few historical episodes to draw on. The Fed is preparing to launch QE2 amid widespread skepticism over the chances of its success, just like the last round (QE1) arguably failed to propel the economy to an ‘escape velocity’. Of course the counterfactual - of no introduction of QE in early 2009 - could have been a much more severe and protracted economic slump.
It is often argued that QE2 will not be successful because at best it will drive down risk free rates a little bit but ten year bond yields have already fallen sharply since April - from roughly 4% to 2 ½% - and housing is still in the doldrums. Moreover, after the initial QE announcement in March 2009, the decline in bond yields proved short-lived. It is also argued that the counterpart to Fed asset purchases – bank reserves on the asset side of the balance sheet – will just sit there and not be lent out into the real economy because the money multiplier is broken. Some commentators therefore conclude that the Fed is effectively ‘out of bullets’ and there is nothing that can be done to avert the inevitable ‘Japanification’ of the US economy. Not even QE3, 4, 5, 6, 7 & 8 will prevent this outcome. Others have warned that quantitative easing – allied to record peace-time fiscal deficits – is ‘the road to hell’ – which could unleash Weimar-style hyperinflation, financial Armageddon and the demise of our fiat money regime. Another possibility, is that quantitative easing just might work and accelerate the transmission of monetary policy, thus pushing the economy on to a sustainable growth track without runaway inflation to meet the Fed’s dual mandate of price stability and maximum sustainable employment in the medium term. This is the outcome the Fed is hoping for and this in our view is what would constitute a successful QE2.
Given that the ‘natural rate’ of unemployment is estimated to be about 4% points below the current unemployment rate of 9.6% it will probably be several years before the employment objective is met. Likewise, while the Fed’s preferred measure of core inflation at around 1% is about 1% point below where they want it, in a normal cycle inflation is the most lagging of indicators and so it could be some considerable time before the inflation objective is met too. I would argue that the key transmission mechanism of QE is through ‘risky assets’ or looser financial conditions, the most conspicuous of which is the stock market, and this is the Fed’s implicit intermediate target. Put differently the Fed will act to put a floor under risky asset prices at least until it is sure that the recovery has plenty of self sustaining momentum. This means that the Fed will not want to risk a significant tightening in financial conditions on any foreseeable market horizon. This is the Bernanke put writ large.
If the economy does not gain traction Fed is likely to to be aggressive in administering ever more doses of QE, and the more this happens the greater the risk of adverse unintended consequences (discussed below) and the more fraught the exit. When a central bank owns a large proportion of a given securities market the notion that it can exit without causing major market dislocation isrisible. A good analogy for the exit would be a Japanese TV game show where the contestant (Bernanke-san?has to negotiate an absurdly contrived obstacle course and avoid falling into the water. It can be done but the odds are heavily stacked againsthim. What is alsoominous is the sense of group think at the Fed and the firm conviction that it is doing the right thing - when if they were really honest about it they would concede that they have no idea what will happen - and that it is the ECB, which is extremely reluctant to go down this route, which is misguided. History is littered with the casualties of politicians and policymakers who have felt the heavy hand of history on their shoulder and let it go to their head. More than any other policymaker in history the spotlight is on Ben Bernanke. He knows his time has come and he is a man on a mission. Will he be able to save the world and be canonized? Or, could he go down in history as the architect of economic chaos? What is QE?
Quantitative easing shifts the focus from the price of money to the quantity of money. QE has been invoked when nominal interest rates have reached, or are close to, the ‘zero bound’. It involves ‘large scale asset purchases” (LSAP), which expand the central bank’s balance sheet. The counterpart to the increase in the asset side of the balance sheet is an increase in reserve liabilities, which in turn increases ‘high powered money’ or the ‘monetary base’. In principal, these reserves could multiply into faster ‘broad money’ growth if banks choose to increase their lending to the real economy. Alternatively banks may prefer to leave their reserves parked at the central bank (we will discuss this issue in more detail below). In QE, the central bank purchases financial assets – usually government bonds, mortgage-backed securities and corporate bonds, but not (yet) equities - from the private sector by creating new money. If the purchase is from a bank the central bank merely credits the reserve account of the bank in question. If it is from a non-bank the central bank credits the account of the seller’s bank, which in turn credits the deposit account of the seller. Money has been created out of nothing. It is an electronic transaction and is not physically printing bank notes, but it amounts to the same thing. Asset purchases from banks increase the monetary base, while asset purchases from non-banks increase both the monetary base and broad money.
The Bank of Japan (BoJ) grudgingly engaged in quantitative easing in the early 2000s sixyears after inflation first turned negative), while other central banks – notably the Fed, the Bank of England and the Swiss National Bank – embarked on their own QE programs in November 2008 and March 2009, to pre-empt the risk of deflation Headline inflation was negative because of the collapse in commodities prices but core was significantly +ve and closer to 2%There are two key pillars of QE. One is to increase the supply of money in the system to support the macro economy, while the other is to address dysfunction in individual markets, often manifested in a sharp rise in illiquidity premia. The moniker of ‘credit easing’ (CE) could perhaps be more aptly applied to the second pillar, which typically involves much smaller amounts. The Fed adopted a wide variety of specific CE programs from late 2007 more than a year before embarking on full blown QE in March 2009. The Fed’s first QE program (QE1) focused on boosting the economy via the housing market and addressing dysfunction in the mortgage market. The Fed purchased $1.45trillion of agency and mortgage backed securities and $300bn of US treasuries (equivalent to 12% of annual GDP over 6-12 months), and therefore QE1 was arguably a bit of a hybrid QE/CE program. The Bank of England adopted a pure unadulterated QE via massive gilts purchases, taking out one year’s entire gross issuance (14% of GDP). The BoE also intervened by launching a commercial paper and corporate bond facility to address dysfunction in these markets. The SNB adopted an entirely different approach by printing Swiss francs to purchase foreign currency, predominantly euro. The total intervention amounted to a staggering 35% of annual GDP to the summer of 2010. QE transmission
The rather simplistic view that QE transmission is through ‘risk free’ assets (government bonds) should turn out to be a sideshow compared to the transmission through other ‘risky assets’ and the US dollar. This is where QE is most likely to gain traction and perhaps even provide that magic spark, which re-energizes animal spirits and lifts the economy into a normal cyclical recovery of sustained above trend growth
The direct impact of long term interest rates is a sideshow for two reasons. First, if a 150bp decline between April and August has so far had no discernible impact on economy and the housing market in particular, another 25-50bp is unlikely to make much difference either. Second, what ultimately matters most for long term interest rates is growth in the economy, so perversely if QE2 is going to work it will eventually be associated with an increase, not a decrease, in long term interest rates. What we can say with more conviction is that QE is likely to push down real interest rates – even if the impact of nominal yields is uncertain – by boosting, or putting a floor under, inflation expectations. A stylized transmission mechanism for asset purchases
Source: Bank of England Quarterly Bulletin, Q2 2009
The transmission through other channels – the indirect impact of lower risk free rates - is arguably more. When the central bank purchases assets from a financial company this tends to increase the liquidity of private sector balance sheets. If the assets exchanged are not perfect substitutes the company will have excess money balances, and the company will seek to rebalance its portfolio by buying other assets. That in turn transfers the excess money balances to the sellers of these other assets, who will look to purchase other assets as well; and so on. This process should bid up asset prices to the point where the share of money relative to assets has returned to its desired level. This is known as the ‘portfolio balance effect.’
Stock effect of LSAP on nominal treasury yield curve Source: Federal Reserve
In addition the relative price of the asset purchased by the central bank – e.g., government bonds – will tend to rise – and their yield will therefore fall - relative to other assets, and this will encourage a substitution into other (riskier) assets as investors look for a higher return. This will push up asset prices more generally, bring down borrowing costs for households and companies and tend to boost confidence and spending. The chart below shows that Fed staff economists’ estimate that the Fed’s treasury purchases drove yields down by around 50bp. The Bank of England estimates that its asset purchases have driven down gilt yields by round 1% point relative to where they would otherwise have been. QE also operates through the expectations channel by underscoring that the central bank is prepared to do whatever it takes to boost the economy and avert deflation (this could also have some adverse unintended consequences discussed below). QE also tends to weaken your currency – at the time of writing the dollar was falling across the board in anticipation of QE2 – for the simple reason that if you increase the supply of your currency relative to other currencies it is likely to go down. Other things equal a weaker dollar will stimulate the economy. But other things are not equal and an unchecked decline in the dollar is likely to have some adverse unintended consequences (see below). Although the US is a relatively closed economy the trade share of GDP has been rising and is now 28% (exports 12%, imports 16%) so swings in the dollar do matter for growth and inflation more than they used to. Moreover at a time when the FOMC is fretting over an unwelcome decline in inflation expectations a weaker dollar can come in handy. However the FOMC should be careful what it wishes for and rightly generally refrains from commenting on the dollar. The other channel is bank lending. Banks will end up with more reserves on their balance sheet as a result of asset purchases. Other things equal banks should be more prepared to hold a higher stock of illiquid assets – loans to the non-bank private sector – than they otherwise would have done. In theory LSAPs could have a huge effect on the economy if they multiply quickly into stronger broad money growth. The standard retort of course is that these reserves will just sit there because the money multiplier is broken and we are in a ‘liquidity trap.’ This view has almost taken on conventional wisdom.
Three points can be made in response to these refrains. First the growth in credit is a function of the supply of credit and the demand for credit. The supply of credit depends – among other things - on the financial health of lending institutions and the demand for credit depends – among other things - on confidence and the level of credit sensitive spending (e.g., demand for housing). Second, positive credit growth is not a necessary condition for economic growth, especially in the early stages of the cycle. Third, and related to the first two points, it is unrealistic to expect the money multiplier to come back immediately. It will take time. To claim that the money multiplier is permanently broken is at best premature. In the US banks are in a position to step up lending and we have seen this in an easing in lending standards – reliably a bullish omen for the economy. Bank lending is being constrained by weak demand, in turn reflecting low confidence. However, while not increasing yet, lending is starting to level off, after falling for two years. This does not mean that there are no new loans being extended, rather it is a reflection of the low level of activity in credit sensitive parts of the economy like housing and autos. New loans extended are being offset by amortizations and write-offs. Gradually, as activity picks up in these sectors, bank lending is likely to turn positive. The base case is that lending picks up slowly from here and gathers pace as the recovery gains more traction but it is likely to remain below the growth of incomes for some time. The tail risks are slightly more skewed to the economy catching fire and lending surprising to the upside, rather than the other way. Unintended consequences
There is an outside chance that QE works too well before the ‘exit strategy’ has been worked out and the economy grows too quickly and inflation picks up. There is an interesting debate, yet to be resolved, as to whether it is the ‘stock’ or ‘flow’ of asset purchases that matters. From the perspective of abundant excess reserves and base money multiplying out it is clearly the stock of QE that matters. If the money multiplier is not permanently broken then QE2 could be ‘over egging the pudding’ and could in hindsight prove to be a serious policy error.
The ‘shock and awe’ nature of QE1 may well have stabilized an economy in freefall and one could think of some grizzly counterfactual outcomes had it not been launched. Today, in marked contrast, the landscape is very different because the economy is growing – albeit at a sedate pace –the expansion is over a year old and the risk of a double-dip recession is remote. Were it not for the Fed’s ‘dual mandate’ – of maximum employment and price stability - the Fed may not even be contemplating QE2 now. In most other countries the former objective is subordinate to the latter. The dual mandate may oblige the Fed to over-stimulate an economy that really doesn’t need any more steroids. By linking the two objectives in the mandate the Fed is implying that there is a permanent trade-off between inflation and unemployment at very low rates of inflation. This has all sorts of implications, one of which could be that the optimal rate of inflation could even be north of 2%.
Commodity prices have been on a tear ever since the markets got wind of QE2, and this is perhaps the most conspicuous adverse unintended consequence, which will squeeze real income growth in the US and other economies. In the last three months sugar prices have risen by nearly 60%, corn and wheat are up by 30% and rice is up 26%. It is not inconceivable that this could even contribute to social unrest in some poorer developing countries. QE2 is also likely to exacerbate overheating and local asset bubbles in other economies, especially those emerging markets that effectively import US monetary policy via managed currency regimes. Not only could this destabilize emerging market economies and exacerbate boom-bust cycles it could increase systemic risks to the global financial system. Given the number of currencies with explicit or de facto links to the dollar the imminence of QE2 has quickly transformed the environment towards competitive devaluations, where no country wants a stronger currency and many are acting aggressively to prevent it. If unchecked, this could degenerate into trade wars and capital controls - an unambiguously negative outcome. By its nature QE tends to be introduced to boost a weak economy, which means that foreign investors get clobbered first by currency devaluation before locals eventually get whacked by higher inflation, which is one reason why it is politically attractive. There are other more sinister reasons too, which are discussed below.
It is possible that QE2 is too successful in putting a floor under inflation expectations and they become unhinged to the upside. One fairly benign adverse unintended consequence could be that the Fed might be more successful at boosting nominal GDP than real GDP. In other words, partly via the expectations channel, QE is associated with a less favorable split between real GDP and inflation. This appears to have happened in the UK, where QE1 was arguably more successful than it was in the US (see appendix), and where QE2 is also likely to be introduced around the same time as in the US (our guess: early November 2010).
QE, inflation vs. hyperinflation, fiscal deficits and gold
Another related but much more insidious risk is the one that keeps policymakers awake at night: hyperinflation. It is something they would rather not even talk about , while other commentators have ridiculed the notion that we could have hyperinflation when they argue the risks are skewed towards deflation. Hyperinflation can happen suddenly and is inextricably linked to fiscal sustainability. It is a very different animal to the common-or-garden inflation we have experienced in the post-war years, which has reflected, among other things, the interaction of economic overheating, monetary policy errors, wage-price spirals and commodity booms, but have had little to do with fiscal policy or fiscal sustainability.
Hyperinflations have happened when governments have resorted to printing money to cover large budget deficits when the scope to raise taxes or borrow has been restricted, leading to a collapse of confidence in the fiat money regime, which then required faster and faster money printing to command the same real resources as the value of money collapsed. Given record peace time deficits in many developed countries, the alarming impact of demographics on long run fiscal trends, and the resort to quantitative easing, the risk of hyperinflation should not be dismissed out of hand. It is true that hyperinflations are more prevalent in economically or politically unstable countries but that is not always the case. Arguably of all the advanced countries Japan is the most vulnerable to hyperinflation today because it lacks any credible fiscal consolidation strategy and tax revenues have fallen below non-discretionary government outlays. For those who have a hard time reconciling the notion that inflation can take off in an economy with a significant degree of spare capacity, the answer lies in the identity: When confidence in the integrity of the fiat money regime is destroyed people can’t get rid of paper money fast enough and flock to foreign currencies and hard assets, and money velocity, V, skyrockets and P soon follows. The integrity of a fiat currency ultimately rests with the sustainability of the fiscal policy regime, or in essence the credibility of the government’s promise to generate tax revenues in sufficient measure to make good on its debt. After the First World War, when the US was on the gold standard, Germany faced staggeringly punitive war reparations, and its fiat currency was un-backed. The government imposed limits on the extent to which it would meet these huge revenue needs through explicit taxation. Inflation really took off after the French occupied the Ruhr in early 1923, as the government increasingly resorted to financing expenditures by discounting treasury bills with the Reichsbank (printing money). Hyperinflation exploded in Germany because the government could not credibly levy taxes to meet its obligations. For all intents and purposes Germany was fiscally insolvent.
Fast forwarding to today, enormous primary and structural budget deficits in the advanced countries beg the same question: are governments fiscally solvent? The IMF calculates that advanced economy government-debt-to-GDP ratios will rise to 110% by 2015, a 37% point increase from pre-crisis levels. In order to reduce debt to GDP ratios to 60% by 2030, they will need to reduce cyclically adjusted primary balances by 8 ¾% of GDP. History has taught us fiat money panics can happen suddenly out of the blue with no apparent catalyst. It is analogous to a large flock of birds. One bird flies off and the whole flock follows an instant later. It is also worth remembering that every fiat money regime in history has eventually failed because of currency debasement. When the fiat money panic arrives the macro backdrop will be irrelevant, it will just arrive. The question that we have to ask ourselves is how close are we to the end of the current regime? Are we in Act V Scene IV, or further back in the drama? The answer is that we simply don’t know.
The adoption of money supply targeting by the Volcker Fed to crack inflation in the early 1980s, may have saved our fiat money system after gold started going exponential in late 1979 (see chart below). Likewise the successful introduction of inflation targeting in the early 1990s also gave fiat currency an important lease of life and gold became the most unloved investment slumping to a 20 year low in 1999. By luck or good judgement inflation targeting central bankers were remarkably successful at meeting their inflation objectives. An imminent collapse in confidence in our fiat money system is certainly not the central case, but the risk is not zero either. The price of gold has increased by 17% a year in the last decade and at a 35% annualised rate since QE1 was first mooted nearly two years ago. The behaviour of gold would set more alarm bells ringing if it had significantly outperformed the rest of the commodity complex but it generally hasn’t.
If gold and other precious metals started to go up exponentially and decouple from the rest of the complex that would raise more red flags – because in the event of the demise of fiat money it would be better to have your pockets stuffed full of krugerrands than have a herd of live cattle in your backyard. Another sign of trouble would be more frequent and incessant official tirades against gold from central bankers and politicians. Even more ominous still would be outright bans being imposed on holding gold.
The history of gold over 40 years
According to a famous living economist, writing in 1966:
In the absence of the gold standard, there is no way to protect savings from confiscation through inflation. There is no safe store of value. If there were, the government would have to make its holding illegal, as was done in the case of gold. If everyone decided, for example, to convert all his bank deposits to silver or copper or any other good, and thereafter declined to accept checks as payment for goods, bank deposits would lose their purchasing power and government-created bank credit would be worthless as a claim on goods. The financial policy of the welfare state requires that there be no way for the owners of wealth to protect themselves. This is the shabby secret of the welfare statists' tirades against gold. Deficit spending is simply a scheme for the confiscation of wealth. Gold stands in the way of this insidious process. It stands as a protector of property rights. If one grasps this, one has no difficulty in understanding the statists' antagonism toward the gold standard.
That economist was Alan Greenspan.
Conclusion
The odds of QE2 being ‘seen’ to work are better than most commentators believe because the economy may be picking up anyway after a soft patch. However given the extent of private deleveraging that is underway, and the limited scope for fiscal policy to provide anymore impetus, significantly above trend growth (4% plus) would now seem to be a remote possibility. Growth of 3-3.5% is a more likely outcome over the next year, which differs from the consensus (about 2-2.5%). If the consensus is right, more and more QE will be administered and the risk of an accident will increase commensurately. One can be are far more confident in the Fed’s ability to boost inflation than real growth, and therefore a realistic possibility would be reasonably decent nominal GDP growth of 6-7% and rather tepid real growth of 2-3% as we have seen in the wake of the UK’s more aggressive QE1 (see appendix two). The ‘Japanification’ scenario (1% real growth and 0% nominal growth) is at least as remote as the real nightmare scenario of a collapse in confidence in the internal purchasing power of the dollar. Ironically Japanification would make the hyperinflation end game more likely because the fiscal position would become unsustainable. When it comes to the risk of hyperinflation it is Japan not the US that is at the front of the queue because Japan’s fiscal position is no longer sustainable: it cannot credibly be rectified with traditional tax increases and spending cuts.
Appendix one: the Fed and the ECB – different philosophies towards quantitative easing
It has become clear from discussions with policymakers that the Fed and the ECB have an almost diametrically opposed view of the virtues of quantitative easing. Moreover, in private it appears that each institution is skeptical, bordering on disdainful, of the other’s approach. While the FOMC itself is divided and the hawks have been unusually vocal both inside and outside closed door meetings, it is arguably only the rump of the FOMC led by Bernanke that really matters. They have all studied the Great Depression and the Japanese experience and they are convinced that the have to be bold and front load policy stimulus….and keep at it until they get results. In private they see the ECB dithering while Rome burns. The ECB just ‘doesn’t get it’. The ECB sees things entirely differently and has a different philosophical approach. ECB policymakers are privately aghast at what they regard as US policymakers’ impatience, particularly because growth is rarely vigorous coming out of a balance sheet recession and there is nothing monetary policy can do about it. Excessive policy activism just creates uncertainty and instability. Moreover the ECB pins the blame for the recent recession on overly lax (US) monetary policy which led to credit booms and created asset bubbles and inevitable bust and deleveraging. Therefore further monetary policy activism will just stoke up more asset bubbles in the future. As we have argued before these philosophical differences also reflect the fact that depression is the nemesis of the US and hyperinflation for Europe. This entrenched philosophical divide could be reflected in continued strengthening of the euro which will frustrate the ECB’s desire to normalize policy.
Appendix two: Comparing the success of QE1 in the US and the UK
As discussed above QE1 in the US was more of a hybrid QE/CE (credit easing) program, while the Bank of England adopted an unadulterated version of QE. The BoE QE program was slightly more aggressive (14% of GDP vs. 12% of GDP in the US) and was also more explicit in its objective of boosting money GDP and spending. Based on this criterion QE1 was conspicuously more successful in the UK. In the US nominal GDP growth has barely exceeded 4%, which must be a disappointment for the Fed. No wonder Bernanke believes that, in hindsight, there was insufficient shock and awe in QE1.
Nominal GDP growth (two quarter annualized rate)
While the Bank of England managed to crank up nominal GDP impressively this has mainly been because of higher inflation, while real growth has been quite pedestrian, and until recently was running below the US. The strength of the UK GDP deflator is echoed in UK CPI data which has been running around 3% on a headline and underlying basis in stark contrast to the US and other developed countries where inflation is running closer to 1%. The US GDP deflator fell been well below 1% for a year on a year on year basis, and has only begun to pick up more recently towards 1 ½% on a two quarter annualized basis, well below the UK’s 4%. GDP deflator (two quarter annualized rate)
Real GDP growth: two quarter annualized rate
In the US case when the Fed introduced credit easing in late 2007 it merely changed the composition of its balance sheet, leaving its overall size unchanged, while in the UK’s case the BoE began to purchase private assets before announcing QE, funded by the issuance of treasury bills rather than central bank money.
