|
文/Hervé Hannoun（国际清算银行副总经理）
I. The build-up of the debt overhang and its causes
A. The long-term rise in debt
Let me start with the big build-up. Graph 1 shows that total non-financial debt in advanced economies rose by 67 percentage points, from 212% of GDP in 1999 to 279% in 2014. It is striking how this increase in debt took place against the background of slowing productivity growth. Of this increase more than half (37 percentage points) occurred since the start of the global financial crisis. Graph 1 also shows that, since 2008, total debt has begun to rise at a rapid pace in emerging markets, reaching 157% of GDP in 2014. As the breakdown in the upper panel of Graph 2 shows, total debt is now almost four times GDP in Japan and in a range of two and a half to three times GDP in the majority of advanced economies. Over the last 15 years, the debt build-up in advanced economies was broadly based, involving both public and private debt. While some deleveraging has taken place in the private sector since 2007 – especially in the United States and the United Kingdom – the decline has been more than offset by a large rise in public debt. Moreover, this increase in domestic debt has gone hand in hand with a substantial rise in external debt. On top of this, many advanced economies also face rising age-related liabilities resulting from underfunded pension and health programmes. As Carmen and Vincent Reinhart and Kenneth Rogoff put it, we are confronting not a single but rather a “quadruple debt overhang” problem.
Debt is much lower in EMEs as a group, as shown in the lower panel of Graph 2. But the average masks significant differences. In particular, China’s total debt is now 229% of GDP. In less than six years it has increased 76 percentage points (or about 50% above its 2007 level). China’s corporate debt stands now at over 150% of GDP, above the levels of most advanced economies. Public debt is relatively low at 40% but, taking into account off-balance sheet liabilities likely to materialise, China’s public debt is estimated to be over 50%. In Korea, total non-financial debt is over 220% of GDP, close to that of China and other advanced economies. In most other EMEs, especially in the Asia-Pacific region, debt is much lower, but its pace of accumulation has accelerated post-crisis. Fast growth in debt is a concern, as it is more likely to be associated with excessive concentration of credit risks and misallocation of resources.
B. Unconventional economic policy and the debt-driven growth model
1. Easing bias of monetary policy
Graph 3 illustrates the easing bias of monetary policy. It compares the actual policy rate with that implied by a simple Taylor rule. A clear departure from this rule occurred in the early 2000s: monetary policy became distinctly too accommodative in the United States and – via monetary policy spillovers – in the rest of the world too.
A common objection is that the steady-state “natural” interest rate has fallen since the late 1990s, making invalid any inference based on a simple Taylor rule. The argument is well known. The emergence of large excess saving (or current account surpluses) in Southeast Asian countries, China and oil exporters required a fall in the equilibrium real interest rate to induce a corresponding widening of current account deficits in other countries. Central banks in advanced economies had no choice but to accommodate these structural forces lest they imparted an undue disinflationary bias to their economies.
First, in the run-up to the crisis, the increase in net capital flows was dwarfed by the huge increase in gross financial flows. And most of these gross flows were between advanced economies, not between advanced and emerging markets.
Second, the bulk of the gross capital inflows into the United States originated from the private sector, not the official sector. And most were not used to purchase government debt securities, but riskier and often toxic assets such as asset-backed securities.
Third, half of the gross inflows into the United States came from the United Kingdom and the rest of Europe (which were in deficit or close to balance respectively). China and other Southeast Asian economies only contributed a minor share.
2. Preference of the banking sector for leverage against capital
Along with regulatory gaps, cheap money has induced banks to raise their leverage ratios in the run-up to the global financial crisis. As shown in Graph 5, the leverage of large banks increased by half, from 20 times equity in 2000 to 30 in 2008. Since the onset of the crisis, the implementation of the Basel III regulatory framework has prompted banks to boost their capital positions, thus leading to a gradual fall in their leverage ratio, which is now around 15 times equity. Further progress may be expected, as important parts of Basel III have yet to be completed. These include, in particular, the calibration of the leverage ratio, the review of sovereign risk weights, and the update of the Basel II capital requirements on securitisation.
3. Widespread preference of economic agents for debt against equity
The incentives for corporates to take on more debt continues to be strong in the post-crisis environment. Abetted by ultra-low interest rates, corporate bond issuance has been strong in bothdomestic and offshore markets. In BIS statistics, the outstanding amount of US dollar and eurodenominated bonds issued by borrowers who reside outside the United States and the euro zone has jumped to over $5 trillion in 2014 as monetary policy has helped to push down term premia. At the sametime, in the United States initial public offerings (IPOs) are currently well below their 2007 level (about$238 billion in 2014 against $438 billion in 2007). And share buybacks have also been rising. For example, in the United States share buybacks totalled $533 billion in the 12 months to June 2014, a 26%increase over the comparable 12-month period in 2013 and close to the high level reached in 2007.
4. Expansionary bias in fiscal policies
This “anti-austerity” thinking, however, needs to confront the reality of unsustainable public debt trajectories. In 2010, a BIS working paper presented a set of 30-year projections for the path of the debt/GDP ratio in a dozen major advanced economies. The results are alarming. Let me quote: “In the baseline scenario, debt-to-GDP ratios rise rapidly in the next decade, exceeding 300% of GDP in Japan; 200% in the United Kingdom; and 150% in Belgium, France, Ireland, Greece, Italy and the United States.”10 Since 2010, fiscal positions and the associated debt projections have improved somewhat, but not enough. Without further progress on closing current fiscal deficits and without reforms that curb pension and health expenditure, debt will inexorably continue to rise.Stronger growth is unlikely to come to the rescue. The historical record shows that high growth is seldom associated with high public debt.12 Policymakers should not fool themselves into believing that the debt overhang problem can be solved without fiscal discipline and rigour.
To sum up, with the exception of some deleveraging in the financial sector, total non-financial debt has continued to grow, prompting the simple question that heads up a recent Geneva report on the world economy: “Deleveraging? What deleveraging?”13 Of the two root causes of the global financial crisis,14 only the microeconomic one has been addressed, through regulatory reform (Basel III) that aims at improving banks’ capital buffers, risk management and incentives. The macroeconomic policy causes – the flaws of the debt-driven growth model and the loose policy mix that led to the crisis – have yet to be addressed. How do we get out of this unsustainable debt-driven growth model?
II. Approaches to the resolution of the debt overhang problem
So far, central banks have acted aggressively to prevent debt deflation from materialising. But with the end of the emergency and now that a recovery is under way, the continued surge in debt raises questions about the appropriateness of keeping interest rates at present ultra-low levels. Indeed, the risk of a policy mistake is increasing: a central bank can normalise interest rates too early or too rapidly, thus choking off the recovery and making it harder to deleverage. But it can also wait for too long, leading to a further accumulation of debt, yet greater financial risk-taking, and the further postponement of essential growth-enhancing fiscal and structural reforms. If debt keeps rising against the backdrop of very low interest rates, the cost of any future adjustment may in fact rise, while the risk of debt deflation is simply pushed into the future.
A. The menu of options for solving the debt overhang problem
Governments, not central banks, have the primary responsibility for solving the debt overhang problem. But what can governments do? In a recent paper, Carmen and Vincent Reinhardt and Kenneth Rogoff lay out the menu of options: faster growth, debt restructuring or default, inflation, wealth taxes, financial repression, privatisation and austerity.16 Their focus is on the reduction of public debt, but some of the options (especially regarding growth and inflation) also apply to the reduction of private debt. Let me consider these options in turn.
A second option is default or debt restructuring. This option may be inevitable if governments fail to take action to consolidate public finances or boost trend growth, but should not be viewed as an appropriate solution. It might in fact involve unpredictable costs and wealth redistribution, which in the end may be less acceptable to society than ordinary fiscal consolidation. Any negotiation is likely to be prolonged and the ensuing uncertainty may in the process cripple the financial system.
First, a large one-off wealth tax could, in principle, be used to eliminate a substantial chunk of outstanding public debt. In this sense, it is similar to a default or debt restructuring except that its burden would fall on all wealth holders, not only on bond holders. Yet, a large wealth tax might in practice be difficult to implement. It is more likely that wealth taxes may become part of the regular set of fiscal tools. If designed appropriately, they could also help reduce inequality and make necessary fiscal consolidation more acceptable to the general public.
A fourth option is a surprise surge in inflation. This would not be an acceptable option from a central bank’s perspective. And it would anyway be insufficient. Without imposing any restrictions on financial transactions, interest rates would soon catch up as creditors seek compensation for higher inflation. The easing of the debt burden would therefore be temporary (and inversely related to the maturity of outstanding debt). But the costs could be permanent. The credibility of central banks’ inflation targets might be lost, resulting in higher, more persistent, and more volatile inflation rates.
This option inevitably leads to slower average growth for some time. But it has the merit of being the most reasonable one, recognising that savings have not become the enemy of policymakers. Defaulting on debt or engineering inflation (which is another stealthy way of defaulting) are far worse solutions to the debt overhang problem.
B. Central banks’ current efforts for the resolution of the debt overhang problem
Confronted with the debt overhang problem, major central banks have – rightly – favoured the option of a gradual increase in savings rates among debtors, broadly supporting fiscal consolidation, while trying to stimulate spending by less indebted agents or net savers. More specifically, their implicit strategy has involved three elements:
1. The first is to forcefully reduce debt service ratios, thus boosting debtors’ cash flows and preventing them from cutting their spending too sharply. The reduction in borrowing costs was achieved initially by large cuts in the policy rate, which had a stronger effect in countries with a large share of variable rate debt such as the UK and several euro area members. Subsequently, borrowing costs were pushed down by quantitative easing measures as well as by forward guidance (the promise to keep policy rates low for longer).
2. The second is to reinforce the commitment to keeping inflation as close as possible to the 2% target. For example, the US Federal Reserve has removed any ambiguity about its target by making it explicit at 2%. The Bank of Japan has also announced a target of 2% along with “open-ended” quantitative easing measures. And the ECB has stuck to its “below but close to 2%” target. This has helped keep long-term inflation expectations well anchored at or just below 2%, thereby preventing any risk of a deflationary spiral. Central banks have also rightly rejected academic advice to raise their inflation targets. Central banks cannot be seen as engineering inflation as a way to reduce the real value of the debt. And, as argued above, the long-run costs would largely outweigh any transitory benefits.
3. The third is to support aggregate demand by reducing the long-term interest rate, bypassing the zero lower bound on policy rates through unconventional measures such as large-scale asset purchases and forward guidance. As shown in Graph 6, these measures have helped push term premia into negative territory in 2011–12 in the United States and the euro area; and again in 2014. A negative term premium is unsustainable in the long run as it means that an investor is willing to pay to be exposed to interest rate risk rather than being compensated for this risk. III. Central banks’ dilemmas on the debt overhang problem
Is the current stance for the resolution of the debt overhang problem still warranted? In tackling this question, central banks need to confront at least five dilemmas.
1. Are unconventional monetary policies incentivising more debt?
The first dilemma is whether a prolonged period of unconventional monetary policies is incentivising more debt and is therefore beginning to do more harm than good. Instead of helping debtors to deleverage, such policies may have become an encouragement to increase leverage further. As the latest BIS annual report puts it, “Low interest rates can also have the perverse effect of incentivising borrowers to take on even more debt, making an eventual rise in rates even more costly if debt continues to grow… Low interest rates do not solve the problem of high debt.” In some countries, negative interest rates on short-term government securities mean that governments now obtain some remuneration from borrowing short term instead of bearing a cost for it. 2. Should central banks insure against the risks of debt-deflation?
The second dilemma for central banks is about the risk of debt deflation. Should central banks insure against this risk? Let me divide this dilemma into two separate questions. First is the question of evidence. Second is the question of how to respond.
Turning first to the question of evidence, it is not clear from the data that there has been a significant rise in the risk of deflation. As shown in Graph 7, headline inflation in major economies has been falling steadily since its 2011 peak, but the recent decline is largely explained by falling commodity prices, which are known to be volatile. Underlying inflation (“core inflation”) has been more stable, especially in the last few months. Wage inflation is generally subdued but stable. If anything, a fall in inflation should support real disposable income and hence growth. 3. Are unconventional monetary policies causing distributional problems and greater inequality?
The third dilemma for central banks is how they can keep their independence if their policies are perceived – rightly or wrongly – as permanently altering the distribution of income and wealth in favour of some social groups to the detriment of others.
One thing, though, is certain. If these transfers go on for longer, and there is a perception in the public that the main effect of quantitative easing is to engineer asset price inflation, questions could be raised about the potential impact of unconventional monetary policies on inequality. In the end, monetary policy might end up being seen as no different from fiscal policy, in which case demands for political control would naturally grow. And, if this were to happen, central banks’ future ability to effectively conduct stabilisation policy could be impaired.
4. Has the threat of fiscal dominance increased?
The fourth dilemma for central banks is how to help with the resolution of the debt overhang crisis while avoiding the risk of fiscal dominance. Policymakers in advanced economies are under pressure from both financial markets literature and some academic circles to search for easy ways to deal with the debt overhang problem. If unconventional monetary policies disappoint, we can expect a proliferation of “new” ideas on how to circumvent the problem, including "helicopter money" and "overt money financing" (for example, "having the central bank buying government securities which are explicitly non-interest bearing and never redeemable").The trend in the public debate towards more and more eccentric proposals is symptomatic that the economic policy compass has gone adrift.
5. What pace of interest rate normalisation can borrowers reasonably withstand?
If central banks conclude that the potential cost of keeping interest rates at unprecedentedly low levels already outweighs any potential benefit, a final dilemma for central banks is how to normalise them. As the “new normal” for interest rates is highly uncertain, it will have to be discovered in the process of exiting current policies. A “normal” rate should be one that current debtors are able to withstand. At the same time, it should no longer encourage debt accumulation and its attendant risks. Market reaction isalso very hard to predict. For all these reasons, normalisation of policy rates should probably proceed gradually. Yet, the challenge for central banks is to ensure this gradualism is designed so that it does not encourage investors to indulge in additional excessive risk-taking.（完）
文章来源：巴塞尔委员会官网 2014年11月20日（文章有删减，本文仅代表作者观点） 本篇编辑：曾智
|
|
|
This “savings glut hypothesis” is, however, very much overstated. As pointed out by my colleagues at the BIS, it leaves several important facts unexplained:
