and the Devil
adair turner became chairman of Britain's Financial Services Authority just as the global financial crisis struck in 2008, and he played a leading role in redesigning global financial regulation.
Illustrations by James Turner
Published May 2, 2016.
*Princeton University Press 2015. All Rights Reserved.
Sometimes the best ideas in economics hide in plain sight. And to his great credit, Adair Turner, the former chairman of the UK Financial Services Authority who now heads the board of the New York-based Institute for New Economic Thinking, has rediscovered one that is usually consigned to a footnote in macroeconomics textbooks. ¶ Turner's new book, Between Debt and the Devil,* offers a convincing argument for the use of "helicopter drop" money (Milton Friedman's coinage back in 1948) to manage the rubble left by the collapse of the global debt bubble. Most likely, you'll initially be inclined to dismiss Turner's case as fiscal sleight-of-hand, a technique that promises too much gain for too little pain to be realistic. But keep reading, and just as likely your skepticism will morph into dismay that an idea this good could have been ignored for so long. — Peter Passell
Seven years after the 2007-2008 financial crisis, the world's major economies are still suffering its consequences.
Eurozone GDP has not yet returned to pre-crisis levels; unemployment exceeded 10 percent at the end of 2015 and inflation is far below the European Central Bank's close-to-2 percent target. Japan continues to struggle with low growth and relentlessly rising public debt. The United Kingdom has begun to grow and create jobs, but GDP per capita is still below the 2007 level and average real earnings are still some 6-8 percent below the pre-crisis peak.
The U.S. recovery has been more robust, but the employment rate remains far below 2007 levels. Inequality has continued to increase. Across the advanced economies, many people are no longer confident that capitalism will deliver rising prosperity generation after generation.
There could be supply-side explanations for this slowdown in economic growth. Working-age population growth has slowed and has turned negative in Japan. Some economists argue that the attainable rate of productivity growth has also declined. But long-term developments in supply-side factors cannot possibly explain the sudden switch from robust growth in many countries before 2008 to slow or nil growth for seven years thereafter.
Low rates of inflation and nominal GDP growth meanwhile make it clear that inadequate demand has played a major role. For advanced economies to grow in line with potential and with about 2 percent inflation, we need nominal demand to grow at something like 4-5 percent per year. Since 2008, actual nominal domestic demand growth has been less than 3 percent per year in the United Kingdom and the United States, around zero in Japan and less than 0.5 percent in the eurozone. Thus we will never get out of the current malaise, return inflation to target or reduce debt levels unless we increase demand in our economies.
Faced with this malaise, it can seem that all policy levers are ineffective. Indeed, many central bankers are keen to stress the limits to what they can achieve. But inadequate nominal demand is one of very few problems to which there is always an answer. Central banks and governments together can create nominal demand in whatever quantity they choose by creating and spending fiat money. Doing so is considered taboo – a dangerous path toward inflationary perdition. But there is no technical reason such "money finance" should produce excessive inflation, and, by excluding this option, we have caused unnecessary economic harm.
This chapter describes why money finance of fiscal deficits is technically feasible and desirable, and why it may be the only way out of our current problems. Here, I offer some specific ways in which we should now use this potentially powerful tool.
This claim is at odds with the historical track record of most advanced economies."
We Never Run Out of Ammunition
The fundamental reason recovery from the Great Recession has been slow and weak is the debt overhang described in detail elsewhere in the book. Collapsing credit supply played a crucial role in driving economies into recession in 2009. But thereafter, debt overhang was the dominant factor, driving reduced private credit demand.
Excessive private debt creation before the crisis left many households and consumers overleveraged and determined to pay down debt. Reduced private consumption and investment then depressed economic growth, producing large fiscal deficits and a rising public debt-to-GDP ratio. Leverage has not gone away, but simply shifted around the economy, from private to public sectors, or among countries. German deleveraging, for instance, has only been possible because Chinese leverage has soared. Overall, developed-economy leverage, public and private combined, has continued to increase slowly, and total global leverage has increased significantly as emerging economy private credit grows at a fast pace.
Attempted deleveraging has thus depressed economic growth, but no overall deleveraging has actually been achieved. And none of our traditional policy levers seem able to overcome this dilemma.
Is Austerity Inevitable?
After the crisis, fiscal deficits increased substantially as tax revenues fell and social expenditures rose. In the United States, the fiscal deficit rose from 3.2 percent of GDP in 2007 to 13.5 percent in 2009; in the United Kingdom, from 2.9 percent to 11.3 percent. In the eurozone, the aggregate deficit grew from 0.7 percent to 6.3 percent. Those increased deficits helped prevent still deeper recession, providing powerful stimulus to nominal demand in the face of private deleveraging.
Indeed, there is no doubt that if governments run fiscal deficits – spending more than they tax and borrowing the money to cover the difference – the immediate direct effect is increased nominal demand. But in some circumstances, that direct effect can be stymied by the offsetting factors.
If short-term interest rates have already been set by the central bank at an optimal level, increased fiscal deficits will provoke rate increases that slow the economy down. If the increased issue of government bonds produces a rise in long-term interest rates, a similar "crowding out" effect may result. And if taxpayers rationally anticipate that fiscal deficits today mean higher taxes in the future, they may save more today, refusing to spend the tax cuts or reducing their private expenditure by as much as public spending rises (the so-called Ricardian Equivalence effect).
Pre-crisis macroeconomic theory therefore tended to the conclusion that fiscal policy had little potential to stimulate even nominal demand, let alone to produce an increase in real output. The counterargument, powerfully put by Brad DeLong (Berkeley) and Lawrence Summers (Harvard), is that in the special circumstances of post-crisis recession, the offsetting factors do not apply. With central banks determined to keep interest rates close to zero, increased fiscal deficits will not provoke rate increases. There is underemployment and spare capacity, so the direct stimulus effect will produce additional real growth as well as price inflation; fiscal deficits might therefore pay for themselves, generating faster growth in GDP than in the stock of debt, and thus actually reducing the future debt-to-GDP ratio. As a result, rational individuals and companies will not worry about how increased public debt can be repaid.
A strong case can therefore be made that fiscal policy stimulus should have been deployed even more aggressively in the aftermath of 2008. Some estimates suggest that United Kingdom GDP was depressed by 3 percent as result of unnecessarily aggressive fiscal consolidation after 2010. And there is no doubt that in the eurozone, where the aggregate fiscal deficit has averaged 1.6 percent between 2008 and 2013 versus 7.2 percent in the United States and 6 percent in the United Kingdom, fiscal austerity has significantly depressed growth.
But the constraints on our ability to use fiscal stimulus must still be recognized. Even if, in some circumstances, incremental fiscal stimulus might reduce future public leverage relative to a no-action alternative, the large deficits actually run up have been accompanied by big increases in public debt-to-GDP – up from 72 percent to 105 percent in the United States, from 51 percent to 91 percent in the United Kingdom, and from 40 percent to 90 percent in Spain, for instance. And while huge Japanese public deficits after 1990 may, as Richard Koo (the chief economist at the Nomura Research Institute) has argued, have helped offset the deflationary impact of private deleveraging, Japan is still left with the question of how its relentlessly rising public debt can be repaid. In the eurozone, fears that rising public debt burdens in peripheral countries might provoke default or eurozone exit did result in rising interest rates, exacerbating the danger that debts would become unsustainable and increasing the cost of credit to the private sector.
Thus there are limits to our ability to use traditional fiscal stimulus to escape the debt trap. Carmen Reinhart and Kenneth Rogoff's analysis suggests that if public debt levels rise above about 90 percent of GDP, adverse consequences for growth are likely to follow. Controversy over their calculations shows that we must not overstate the importance of any one specific threshold. But their overall conclusion that high debt-to-GDP ratios will inevitably constrain the scope for fiscal policy stimulus is valid.
It is important not to misinterpret this finding. It most certainly does not mean that fiscal austerity is costless because of some so-called confidence-inducing effect. Indeed, the best interpretation of Reinhart and Rogoff's empirical results is that the adverse effect on growth that they observe derives primarily from the fiscal tightening that high levels of accumulated debt appear to make necessary.
That makes it crucial to constrain public debt levels in the good years – and also crucial to restrict excessive private credit creation, reducing the danger that excessive debt will shift to the public sector in the aftermath of crisis. But it also means that we need to find ways to stimulate nominal demand that do not result in rising public debt.
Ultra-Loose Monetary Policy and Adverse Side Effects
For seven years, central banks have tried to use ultra-loose monetary policy to stimulate the economy. Short-term interest rates have been close to zero in the United States and the United Kingdom since 2009, in Japan for much longer, and in the eurozone since 2013. Quantitative easing – central bank purchases of government or other bonds – has been used in Japan, the United States and the United Kingdom to drive down long-term interest rates; in March 2015, it was also finally deployed in the eurozone. And central bank liquidity and funding schemes – such as the Bank of England's Funding for Lending Scheme and the ECB's Targeted Long-term Repo Operation – have sought to ensure that real economy households and businesses, as well as financial market traders and investors, can borrow at low interest rates.
Those policies have almost certainly generated faster nominal demand growth than would otherwise have occurred and helped prevent either still-lower inflation or still-lower real growth. But they have suffered from two deficiencies. First, they have proved insufficient to deliver robust growth, with recovery still anemic and inflation falling below target in all major economies. Still lower (that is, negative) rates would have delivered more stimulus, but if central banks set rates at a more than marginally negative level, individuals and companies would convert bank deposits into currency notes and the stimulus effect would be undone.
Kenneth Rogoff has argued that we should overcome this problem by abolishing paper currency, with all money held in deposit form. Central banks could then set interest rates at significantly negative levels. But that option is not available today. And if it were available and used, it would exacerbate the second deficiency of ultra-loose monetary policy – its adverse side effects.
Quantitative easing works because low long-term yields drive up asset prices and wealth, and thus stimulate asset holders to consume or invest more. It is therefore bound to increase inequality. Sustained ultra-low interest rates, meanwhile, are likely to encourage risky and highly leveraged financial speculation long before they stimulate real demand. And they can only stimulate real demand by encouraging a return to the private credit growth that first created the debt overhang problem. The United Kingdom's Office of Budget Responsibility forecasts that UK private leverage, having declined slightly over the past five years, will by 2020 have risen to its highest-ever level.
The IMF was therefore right to warn in its October 2014 Global Financial Stability Report that "the extended period of monetary accommodation and the accompanying search for yield is leading to credit mispricing and asset price pressures and increasing the danger that financial stability risks could derail the recovery." But in its simultaneously published World Economic Outlook, the IMF also warned that increased nominal demand is needed and that "in advanced economies, this will require continued support from monetary policy." With fiscal policy blocked, ultra-loose monetary policy thus seems both dangerous and essential. Fortunately, however, there is an alternative.
Helicopter Money with Fractional Reserve Banking
Milton Friedman explained most clearly why inadequate nominal demand is one problem to which there is always a possible solution. If an economy was suffering from deficient demand, he suggested, the government should print dollar bills and scatter them from a helicopter. People would pick them up and spend them. Nominal GDP would increase, and some mix of higher inflation and higher real output would result.
The precise impact of any given size of helicopter-money drop would depend on how much people spent rather than saved of their newfound financial wealth. But it would clearly be somewhat proportional to the value of bills printed and dropped. If they were only worth a few percentage points of current nominal GDP, the stimulus to either real growth or inflation would be quite small. If the currency were worth many times nominal GDP, the effect would be large and primarily take the form of increased inflation, since the potential for real output growth is constrained by supply factors.
While Friedman's example is very simple, it illustrates three crucial truths. We can always stimulate nominal demand by printing fiat money. If we print too much, we will generate harmful inflation; but if we print only a small amount, we will produce only small and potentially desirable effects.
The money drop from Friedman's helicopter is fiat money in currency note form – actual dollar bills. And there are historical examples of governments that used printed currency to stimulate nominal demand without generating dangerously high inflation. The Pennsylvania colony did so in the 1720s, and the Union government paid its soldiers with printed greenbacks in the American Civil War. However, most money today is held as bank deposits, not paper currency.
But the essential principle of the helicopter money drop can be applied in the modern environment. A government could, for instance, pay $1,000 to all citizens by electronic transfer to their commercial bank deposit accounts. (Alternatively, it could cut tax rates or increase public expenditure.) The commercial banks, in turn, would be credited with additional reserves at the central bank, and the central bank would be credited with a money asset – a perpetual non-interest-bearing bond due from the government. The "drop" is of electronic accounting entries rather than actual dollar bills, but the operation is in essence the same – as would be the first round impact on nominal demand. The extent of that stimulus would be broadly proportional to the value of new money created.
Printing money in its modern electronic form is thus without doubt a technically possible alternative to either pure fiscal or pure monetary policy. It is, indeed, essentially a fusion of the two. It entails monetary finance of an increased fiscal deficit, and it would stimulate demand with more certainty and with less adverse side effects than either pure fiscal or pure monetary policy.
Compared with funded fiscal stimulus, it is bound to be more stimulative since there is no danger of either crowding-out or Ricardian Equivalence effects. As Ben Bernanke put it in 2003, if consumers and businesses received a money-financed tax cut, they would certainly spend some of their windfall gain since "no current or future debt servicing burden has been created to imply future taxes." And compared with a pure monetary stimulus, it works through putting new spending power directly into the hands of a broad swath of households and businesses, rather than working through the indirect transmission mechanism of higher asset prices and induced private credit expansion. It does not rely on regenerating potentially harmful private credit growth. Nor does it commit us to maintaining ultra-low interest rates for a sustained period.
Our technical ability to stimulate nominal growth with money-financed deficits is therefore not in any doubt. A formal mathematical paper by Willem Buiter, the chief economist at Citigroup, confirms the commonsense arguments of Friedman and Bernanke. His paper title: "The Simple Analytics of Helicopter Money: Why It Works – Always."
Indeed, the crucial issue is not whether money-financed fiscal deficits are feasible and potentially beneficial in the short term, but whether we can contain their long-term impact in a modern economy with fractional reserve banks. For though the first-round impact of an electronic deposit drop is determined simply by its size, the exercise creates additional commercial bank reserves at the central bank and thus makes it easier for banks subsequently to create additional private credit, money and spending power.
Thus the danger exists that the initial stimulative effect of money finance will be harmfully multiplied by subsequent private credit creation, producing more demand stimulus than desired. That danger would not arise in a system of 100 percent reserve banks of the sort that Irving Fisher and Henry Simons supported in the 1930s, and that Milton Friedman recommended in 1948. In such a system, the monetary base is the money supply, private credit and money creation play no role and the final long-term stimulative effect of money finance is bound to be broadly proportionate to its initial size.
For Fisher, Simons and Friedman, 100 percent reserve banks and overt money finance of small fiscal deficits were thus a logically linked package. The latter made it unnecessary to rely on unstable private credit creation to grow nominal demand; the former both made private credit creation impossible and ensured that the long-term stimulative effect of money finance could be precisely controlled.
Fractional reserve banks complicate the implementation of money finance. But the model of 100 percent reserve banks also suggests the obvious solution: any dangers of excessive long-term demand stimulus can be offset if central banks impose reserve asset requirements. These requirements would force banks to hold a stipulated percentage of their total liabilities at the central bank and thus would constrain the banks' ability to create additional private credit and money. Those ratios could be imposed on a discretionary basis, with the central bank increasing them if inflation threatened to move above target. But they could also in theory be deployed in an immediate and rule-driven fashion, increasing the required reserves of commercial banks at the same time as the electronic money drop and by precisely the same amount.
This would essentially impose a 100 percent reserve requirement on the new fiat money creation. We can, in effect, treat the banking system as if it were in part a 100 percent reserve system and in part a fractional reserve: we do not have to make an absolute either-or choice.
The precise consequences of reserve requirements would also depend on whether the central bank paid interest on them, and at what rate. Central banks can choose to pay whatever rate they want on required reserves, but the rate would have to be zero on at least some reserves to ensure that money finance today did not result in an interest expense for the central bank in the future or in central bank losses that would need to be paid for by government subsidy and ultimately by taxpayers.
Setting a zero interest rate for reserve remuneration might, in turn, seem to impair the central bank's ability to use reserve remuneration as a tool to bring market interest rates in line with its policy objective. But central banks could overcome this problem, for instance, by paying zero interest rates on some reserves, while still paying the policy rate at the margin.
Reserve requirements remunerated at a zero interest rate in turn effectively impose a tax on future credit creation. But taxing credit creation might be a positively good thing. Our challenge is to find a policy mix that gets us out of the debt overhang created by past excessive credit creation without relying on new credit growth. Money-financed deficits today plus implicit taxes on credit intermediation tomorrow might well be the optimal combination.
Friedman was right: governments and central banks together can always overcome deficient nominal demand by printing and spending money. That is just as well since, without the option of money finance, there may be no good way out of our debt overhang predicament.
Deleveraging – No Other Good Way Out?
Total economy-wide leverage in advanced economies, public and private combined, is now at levels only previously seen in the after-math of major wars. Analysis of how deleveraging from previous peaks was achieved illustrates just how difficult it will be from today's levels.
The United Kingdom came out of the Second World War with public debts of 250 percent of GDP, but was able to reduce these to 50 percent by 1970. That reduction was not achieved by paying down absolute debt levels; instead, it resulted from 25 years in which nominal GDP grew at about 7 percent per year, while interest rates averaged much less.
That nominal GDP growth rate, in turn, reflected both average inflation of more than 4 percent (well above current central bank targets) and a real growth rate of almost 3 percent, made possible both by significant demographic expansion and by technological catch-up toward U.S. levels of productivity. Moreover, a falling public debt ratio was accompanied by private debts rising slowly from low levels and constrained by quantitative credit controls.
Residential mortgages were only provided by building societies (mutual savings and loans institutions), not by banks, and consumer credit availability was limited by rules on minimum down payments and payback periods. In 1964, total private-sector bank debts, household and company combined, were still just 27 percent of GDP rather than the 120 percent reached by 2007.
The same pattern of rapid nominal GDP growth, low interest rates and low-but-rising private leverage also lay behind the United States' success in cutting public debt from 120 percent of GDP in 1945 to 35 percent in 1970. Across continental Europe, meanwhile, wartime debts were in many countries eroded by high inflation or debt write-off in the immediate aftermath of the war.
The historical experience thus illustrates that public deleveraging is possible, but it also indicates how difficult simultaneous public and private deleveraging will be in today's changed circumstances. Demographic and technological factors will not allow the real growth rates observed in many advanced economies in the 1950s and 1960s. And if 2 percent inflation targets are considered sacrosanct, nominal GDP growth in many advanced economies is unlikely to exceed 4 percent. In some, it will be lower still: the Bank of Japan estimates that Japan's potential growth rate is no more than 1 percent. Thus, even if it achieves its 2 percent inflation target, nominal GDP will grow at only 3 percent.
Growing out of debt burdens will be far more difficult than in the post-war period. Indeed, in some countries the mathematics make it impossible. The IMF Fiscal Monitor illustrates that Japan would need to turn today's primary deficit of 6.0 percent (that is, its fiscal deficit before interest expense) into a surplus of 5.6 percent by 2020 and to maintain that surplus for an entire decade to reduce net public debt to 80 percent of GDP by 2030. This will not occur; if attempted it would push Japan into a deep deflationary depression in which public debt leverage, far from falling, would almost certainly rise.
Japanese government debt will simply not be repaid in the normal sense of the word. Italy's public debt burden, at 132 percent of GDP and rising, is also now so high and the country's potential long-term growth so low that there is no clear austerity-plus-growth path to fiscal sustainability.
Indeed, across the eurozone the Fiscal Compact requirement that all countries should reduce their debt stocks to a maximum of 60 percent of GDP by running primary budget surpluses is not credible. To achieve this objective, Greece would have to run a primary budget surplus of 7 percent of GDP for more than a decade; Ireland, Italy and Portugal would require 5 percent surpluses and Spain 4 percent. As Barry Eichengreen of the University of California has pointed out, there are close to no historical examples of such large continued primary surpluses.
They could only be compatible with robust growth if offset by rapid and potentially dangerous private credit growth either within the countries involved or in their export markets. And if, as is more likely, they produced low growth and sustained high unemployment, debt burdens would not, in fact, be reduced. In the face of such austerity, moreover, talented young people would be likely to leave their countries, reducing the tax base and walking away from their share of the inherited debts.
Sometimes debts simply cannot and will not be fully repaid. Other ways out of the debt overhang will have to be found.
Inflation and Financial Repression
As the UK post-war experience suggests, one option might be to accept many years or even decades in which interest rates are held below nominal GDP growth rates and probably below inflation. One variant of this policy approach – floated by both Kenneth Rogoff and Olivier Blanchard – would entail accepting a higher inflation target than today's 2 percent. Another would be to sustain interest rates close to zero for many more years.
But, in essence, this policy would simply be a continuation of today's ultra-loose monetary approach, reflecting a realistic assessment that it can only erode debt burdens significantly if maintained for far longer than currently hoped. It would therefore suffer from the disadvantages already discussed. It would help erode the value of existing debts, but could only do so by stimulating new credit growth, and it would create incentives for risky financial speculation.
Default and Debt Write-Off
If debts cannot be eroded away by either real growth or inflation, they could be reduced by default and debt restructuring. Rather than creditors receiving an undiminished nominal value degraded in real terms through inflation, the nominal value of debts could be reduced. Such debt write-offs could certainly play a useful role, but they cannot be a sufficient solution.
The extreme version of this option suggests that all we need is fiscal, monetary and free-market discipline. Governments should make their own debts sustainable by cutting expenditures or raising taxes; interest rates should return to normal levels. And in the face of subsequent recession, individuals, companies and governments unable to pay their debts should default, providing a useful signal to creditors to be more careful about lending money in the future.
This policy is essentially the one proposed by U.S. Treasury Secretary Andrew Mellon in 1931: "liquidate labor, liquidate stocks, liquidate farms, liquidate real estate. … It will purge the rottenness out the system." And its consequences would be similar to those that followed in the early 1930s. For as Irving Fisher described in his theory of the "debt deflation" cycle, default and bankruptcy on a large scale drive a self-reinforcing cycle of collapsing nominal demand, as bankruptcy provokes fire sale reductions in asset prices, and as creditors facing unexpected losses themselves cut consumption and investment. The policy of applying pure free market discipline amounts, indeed, to a rejection of the consensus that it is desirable by one means or another to achieve a slowly growing level of nominal demand.
The more realistic alternative involves negotiated debt write-downs and restructurings to reduce debts to sustainable levels while avoiding the disruptive effect of bankruptcy and default. It can be applied to either private or public debts. But in neither sector can debt restructuring alone be sufficient to cope with the scale of today's debt overhang.
Atif Mian (Princeton) and Amir Sufi (Chicago) argue that the United States should have implemented a large-scale program of coordinated mortgage debt restructuring after 2008. By cutting mortgage debts to affordable levels, this would have reduced the severity of the household consumption cuts that drove the country into recession.
Even without such a coordinated program, household debt write-offs have been greater in the United States than elsewhere, helping achieve a more rapid pace of household deleveraging. But Mian and Sufi are surely right to argue that a more extensive and officially mandated program of debt forgiveness would have spurred economic recovery.
But achieving sufficient private debt write-down to fix the debt overhang problem is made difficult by the dilemma that even lending that is "good" from a private perspective can have an adverse macro effect. Overleveraged households and companies can act in ways that depress nominal demand even if they can and do repay their debts in full. Indeed, it is the consumption and investment cuts they make to repay their debts that depress the economy.
For private debt restructuring to be a complete solution it would therefore have to involve the write-down of debts that from a private perspective look sustainable. Orchestrating such a resolution in a fair, politically agreed-on and non-disruptive fashion would be extremely difficult.
Public debt write-offs might potentially play a larger role. In 2011, Greek public debt was reduced by a write-down of private-sector claims without significant market disruption, and public-sector claims on Greece could be and almost certainly will be written down as well. Indeed, public debt write-downs can be used as an indirect way to deal with excessive private leverage. Excessive private credit creation produces crisis, debt overhang and post-crisis deflation, and as a result, rising public debt burdens. Leverage doesn't go away, it simply shifts from the private to the public sector. But once it has shifted to public debt, it may be easier to negotiate restructuring and write-down without harmful shocks to confidence.
The absolute size of the write-downs is crucial, however. The restructuring and write-down of Greek government debt was easily absorbed by financial markets because the total value written off was trivial in global terms. Write-downs of Japanese or Italian government debt sufficiently large to make the remaining debt clearly sustainable would be far more disruptive.
A Combination of Levers
Given the scale of the debt overhang created by past credit growth, there are no certain and costless routes to deleveraging and no one policy that will ensure an optimal result. A combination of policy levers is needed in response, varying by country. In Japan it would not be possible to reduce public debt substantially as a percentage of GDP through the normal processes of growth plus fiscal consolidation. In the United States, starting with smaller debt-to-GDP than Japan's and with faster potential growth rate because of a still-growing population, a combination of continued loose monetary policy, growth and market-driven debt write-down may prove sufficient without more radical policy action.
But whatever the mix of policies deployed, the money finance option should not be excluded as taboo. Indeed, in some countries it will be essential if we are to achieve adequate debt reduction and reasonable growth.
Overt Money Finance – Three Specific Options
Three specific uses of overt money finance should be considered: Bernanke's helicopter, one-off debt write-off, and radical bank recapitalization.
In 2003, Ben Bernanke proposed that Japan execute a modern version of a helicopter money drop, paying for either tax cuts or increased public expenditure with central-bank-created fiat money, making it clear that no new fiscal debt had been incurred and thus that no additional debt-servicing burden had been created. If Japan had followed that advice, it would now have higher nominal GDP, some mix of higher real output and a higher price level, and a lower ratio of debt-to-GDP.
Ideally, the major advanced economies would have implemented Bernanke-style helicopter money drops in the immediate aftermath of the 2007–2008 crisis. If we had done so, the recession would not have been so deep, and we would now be further along in escaping the debt overhang. We would also almost certainly be further advanced in returning to normal interest rate levels. In the United Kingdom, for instance, the Bank of England has conducted quantitative easing asset purchases to the tune of £375 billion (about $535 billion). These have stimulated the economy by pushing down long-term interest rates and increasing bond, equity, and property prices. If instead the UK government had devoted a fraction of that money (say, £35 billion or about $50 billion) to tax cuts or expenditure increases funded with permanent fiat money, the likely effect would have been a stronger, more equitable, and less risky recovery.
In the United Kingdom and the United States, the time for such policies may now have passed. For good or ill, we have used ultra-loose monetary policy to achieve at least some economic recovery. But in Japan and the eurozone the case for money finance has become stronger over the past few years, given accumulating evidence of chronically deficient demand.
At the Federal Reserve's annual Jackson Hole conference in August 2014, European Central Bank President Mario Draghi noted that without some fiscal as well as monetary stimulus, recovery in the eurozone could not be assured. But any fiscal stimulus funded with new public debt issues would raise questions about how the debt could be repaid. Ideally, the eurozone should now consider policies of the sort put forward by the Italian economists Francesco Giavazzi and Guido Tabellini, who argue for a simultaneous three-year tax cut in all eurozone economies, funded by long-term bonds, which the ECB would buy and hold in perpetuity. But the political difficulties may well make this ideal policy unattainable in practice.
Public Debt Write-Offs
The Japanese did not follow Bernanke's advice in 2003. Instead they offset private deleveraging with large funded fiscal deficits, making a relentless rise in public debt-to-GDP inevitable. But they could now write off some of that accumulated debt, putting themselves in the position they would have been if they had accepted Bernanke's advice.
In an attempt to counter deflation, the Bank of Japan has conducted very large quantitative easing operations, buying government bonds that by the end of 2014 amounted to 44 percent of GDP. It is now buying government bonds at the rate of ¥80 trillion per year (close to $700 billion), a figure substantially larger than the fiscal deficit and net new debt issue, which is running at about ¥50 trillion (about $420 billion) per year. As a result, the amount of government debt not owned by the Bank of Japan is falling, and by 2017 net government debt owned neither by the Bank nor by other government-related entities (such as the social security fund) could be down to just 65 percent of GDP.
This seems like a form of money finance. But the stated objective of these quantitative easing operations is not to fund the government deficit but to stimulate the economy through the classic transmission mechanisms of loose monetary policy – very low long-term interest rates, rising asset prices and currency depreciation. And the stated intent is that, at some time, the Bank of Japan will sell its government bond holdings back into the market and that the government will repay these bonds out of future fiscal surpluses.
Official figures for Japanese public debt therefore include the debt that the Japanese government owes to the Bank of Japan, an institution that the government owns. That debt could be written off and replaced on the asset side of the Bank of Japan's balance sheet with an accounting entry – a perpetual non-interest-bearing debt owed from the government to the bank.
The immediate impact of this on both the bank's and the government's income would be nil since the interest that the bank currently receives from the government is subsequently returned as dividends to the government as the bank's owner. So in one sense a write-off would simply bring public communication in line with the underlying economic reality. But clear communication of that reality would make it evident to the Japanese people, companies and financial markets that the real public debt burden is significantly less than currently published figures suggest and could therefore have a positive effect on confidence and nominal demand.
The equivalent operation could also be used to cut stated debt levels and to reduce the apparent need for fiscal consolidation elsewhere, too. The Bank of England owns government bonds worth 23 percent of GDP. Writing off some of them would not entirely remove the need for further improvement in public finances, but would reduce the required pace and severity of fiscal consolidation.
The third possible use of fiat money creation would not deal with the inherited debt overhang but would facilitate rapid progress to a sounder financial system without exacerbating the deleveraging problem.
I have argued that bank capital requirements should be set far higher than those established by Basel III. But rapid progress to higher capital ratios could increase the pace of private-sector deleveraging: banks could meet the higher ratios by cutting loans rather than increasing capital. A possible answer is to require banks to increase ratios by raising new capital and to give banks a short period to raise it from the private sector, but with the backstop of government equity injection if private capital is not forthcoming. Government stakes could then be sold off over time.
The problem is that if the government equity injection is actually required, government debt-to-GDP increases. So if public debt is already at troubling levels, we solve one problem but exacerbate another. Concerns about the impact that public recapitalization would have on already-high fiscal deficit and debt levels undermined the effectiveness of the European bank stress tests in 2011. With Spain, Ireland and Italy already struggling with high public debts and increasing government bond yields, they could not promise to put new capital into the banking system for fear of exacerbating market concerns about sovereign debt sustainability. It was impossible to promise a credible public backstop to private equity raising.
If, however, public recapitalization were financed by the central bank through a permanent increase in central bank money, the problem of future debt sustainability would not arise. Even for those who worry a lot about debt monetization, this option might be acceptable. Bernanke's helicopter money drop may be unacceptable if it took the form of tax cuts or public expenditure increases: the medicine might taste so sweet that the temptation to use it to excess would be overwhelming. But a helicopter money drop used solely for bank recapitalization would be more likely to be treated as a one-off.
As Milton Friedman made clear, deficient nominal demand is one economic problem to which there is an obvious and always possible solution in the form of money creation by government fiat. We have tools, moreover, that can ensure that the demand stimulus is appropriately modest rather than dangerously inflationary. If money finance is excluded, escaping the debt overhang will be far more difficult and economic growth unnecessarily depressed.
But using central bank money to finance fiscal deficits or to write off past public debts remains a taboo policy, and for some good reasons. For if we first admit that money finance is possible, how will we ensure we do not use it to excess? The risks of money finance are thus not technical but political – and the subject of another chapter of this book.