The myth of the "global savings glut": Excessive risk-taking caused the crisis, not excess savings

Several speeches by central bank officials these days contain statements describing the need to eliminate the so-called global imbalances (the notion that large trade deficits and trade surpluses in different parts of the world are unsustainable).

Of course, re-balancing trade flows would certainly be a positive development, as it would help to unlock the grip that large surplus countries have on world production and enable other countries to more fully partake in the global economy.  Such a re-balancing might even trigger a dispersion of manufacturing and other commercial activities away from China and other surplus countries toward other emerging markets and, hopefully, back to developing countries.

From the standpoint of global fairness and economic development, the need to re-balance is a justified, sensible proposition.

Also, many people (and policymakers) seem to think that a re-balancing is necessary in order to prevent another financial crisis from occurring. At the heart of this belief lies the idea that it was the savings of countries running large current account surpluses (read China) that brought down US interest rates and led to the credit boom that caused the crisis. This is the story of the "global savings glut", as Ben Bernanke depicts it. In my view, the story makes for an interesting theory; however, according to the facts, the story is closer to myth than reality.

I've already addressed here the argument suggesting that low interest rates were responsible for the crisis. However, there are several other reasons to doubt the global savings glut story, many of which are found in this paper by Claudio Borio and Piti Disyatat of the Bank for International Settlement (BIS) entitled "Global imbalances and the financial crisis: Link or no link?". In the paper, Borio and Disyatat present seven key inconsistencies associated with the claim that savings from China and other countries running current account surpluses had a role in causing the crisis (p. 4).

First, the authors question whether there is a strong link between current account balances and US dollar long-term interest rates. For instance, the paper shows that, while long-term interest rates were increasing between 2005 and 2007, there was no apparent decrease in the US current account deficit.

Second, the authors note that the depreciation of the US dollar during the last decade is inconsistent with the claim that there was increasing demand for US assets from non-residents.

Third, the authors present evidence demonstrating that the link between the US current account deficit and global savings isn't as strong as commonly suggested. For instance, Borio and Disyatat show that, while the deficit began its deterioration in the early 1990s, the world savings rate actually trended downward to the end of 2003. Also, the paper shows that, while the savings rate in emerging markets has been increasing since 2006, the US current account deficit has tended to stabilize or narrow.

Fourth, the authors show that real world long-term interest rates have trended downward since the early 1990s, irrespective of changes in the global savings rate.

Fifth, the authors note that the rise in the savings rate of emerging markets is inconsistent with the strong growth that occurred between 2003 and 2007. If anything, the authors argue, the rise in savings should have instead depressed aggregate demand and slowed global growth.

Sixth, Borio and Disyatat argue that credit-fueled growth was not associated solely with deficit countries. Countries with surpluses, such as Brazil and India, have all had their bouts of credit booms recently.

Lastly, the authors point out that the countries viewed as responsible for the crisis (read again China) were those least affected by the crisis.

While all seven of these explanations are meant to refute the hypothesis of the global savings glut, it is the last two that provide the biggest clues for understanding what really caused the crisis. According to Borio and Disyatat, it's not the large current account deficits but rather the increasing gross capital flows since the 1990s that are to blame. Conceptually, the authors object to the focus on current account balances as an explanatory variable because it fails:

...to distinguish sufficiently clearly between saving and financing. Saving, as defined in the national accounts, is simply income (output) not consumed; financing, a cash-flow concept, is access to purchasing power in the form of an accepted settlement medium (money), including through borrowing. Investment, and expenditures more generally, require financing, not saving. The financial crisis reflected disruptions in financing channels, in borrowing and lending patterns, about which saving and investment flows are largely silent. (p. 1) (original emphasis)

In other words, the problem with using current account statistics, the authors claim, is that they capture the net financial flows that arise from trade in real goods and services but exclude the underlying changes in gross flows and their contributions to existing stocks, including those transactions relating to trade in financial assets consisting of the overwhelming majority of international financial activity. Also, the current account is silent about the extent to which domestic investments are financed from abroad.

Borio and Disyatat demonstrate these points by showing how gross capital flows rose from approximately 5 percent of world GDP in 1998 to over 20 percent in 2007, with the bulk of this expansion reflected in flows between advanced economies despite a decline in their share of world trade (p. 13).

Also, the authors show that the most important source of capital inflows into the US before the crisis originated in Europe, not emerging markets (p. 15). Finally, the paper shows the extent to which European banks were prominent in global banking during the last decade:

Since 2000, the outstanding stock of banks’ foreign claims grew from $10 trillion to a peak of around $34 trillion by end-2007, an expansion that is striking even when scaled by global GDP [...]. European banks accounted for a large fraction of this increase. (p. 16)

As a result, Borio and Disyatat argue that the causes of the financial crisis cannot be properly understood by looking at changes in national savings rates or real economy indicators such as current account balances. The fact that these indicators do not capture the underlying trend toward increased risk-taking on the part of banks and their involvement in supporting the expansion in the US housing market make them meaningless for understanding the true causes of the crisis.

Instead, the authors conclude that a large part of the blame lies with the banks, as well as the inadequacy of the financial system's regulatory regime and its inability to prevent excessive risk-taking and the development of credit and asset price booms.

Borio, C. and P. Disyatat, "Global imbalances and the financial crisis: Link or no link", Bank for International Settlement Working papers No. 346.

The Volcker Years

Good exchanges on interesting topics deserve their rightful place in the world of economic blogging. Hence, I'm creating a direct link to the recent discussion on Paul Volcker and economic policy in the 80s that followed my last post (h/t: Goffredo, for the suggestion). No offense to Miles...

In regard to the specific discussion on whether monetary or fiscal policy was responsible for lowering the unemployment rate and solidifying the recovery in the mid-80s, I've decided to copy the following excerpt from an article by economist James Tobin:

...it was the Fed’s reversal of policy in late summer 1982 that turned the economy around. No doubt the Reagan fiscal stimuli, as they were phased in, gave a big push to aggregate demand. But the Fed nevertheless managed the recovery in 1983–1984, braking it when it seemed too fast, and relaxing the brakes when GNP growth faltered...

...Could we have had the same recovery, the same path of GNP and employment, under the much more moderate fiscal regimes of the pre-Reagan years? I think the answer is clearly “yes”; the Federal Reserve had plenty of room to lower interest rates further and faster. Would it have done so? That is more debatable. Although the Fed’s change of heart in 1982 signaled its willingness to adjust its policy to macroeconomic performance rather than to money supply targets, it is possible that residual monetarist concerns would have prevented the Fed from fully replacing fiscal stimulus absent or withdrawn.("How to think about the deficit", New York Review of Books, September 25, 1986) (my emphasis)

Although Tobin here first seems to give full credit to the Fed for the recovery, he later questions whether monetary policy alone was responsible. Most likely, Tobin concludes, it was the combination of monetary and fiscal policy that reinvigorated the economy in the 80s. A well-nuanced argument indeed.

Also, I thought it was relevant to link the following article by blogger Lord Keynes. It contains an interesting analysis on Volcker's first term at the Fed. I strongly recommend this article, as it captures my own position on the matter and highlights some key points made by (some of my favorite economists) Alfred Eichner and James Galbraith. Here's an excerpt:

What actually happened under Volcker is that his “floating” interest rate policy caused the federal funds rate to soar to 19% by June 1981, inducing two severe recessions, the first from January to July 1980, and the second from July 1981 to November 1982. This caused mass unemployment, crippled American manufacturing enterprises in the Midwest, and a Third World debt crisis (Galbraith 2009: 38), as the American recessions and high interest rates essentially caused a global recession (Eichner 1988: 548). The high interest rates in the US also lead to a damaging appreciation of the US dollar late in 1980, which hit US exporters hard (Eichner 1988: 549).

The recessions, the US demand contraction and steep fall in the price of oil (which can be seen in this graph) were the real reason US inflation fell from a peak of 14.8% in March 1980 to 4.6% in November 1982, and not because money supply growth rates were brought under control in the way imagined by monetarism or the quasi-monetarist targeting Volcker pursued.

Lord Keynes's point regarding the fall in the price of oil is in line with Warren Mosler's take on the Fed's actions in the early 80s. According to Mosler,

Volcker did not crush inflation. If anything, his rates added to business costs and unearned income long after inflation turned down. The positive supply shocks in the energy markets is what broke the back of inflation, led by the deregulation of natural gas in 1978 that did the lion's share of cutting the demand for crude for electricity generation.

Finally, I'm also including a link to Martin Feldstein's brilliant American Economic Policy in the 80s, published in 1995. The book contains excellent articles on that period, including an article and commentary by Michael Mussa, Paul Volcker and James Tobin on monetary policy in the 80s. Also, I'm linking an article by John Kenneth Galbraith on his preferred approach for dealing with inflation, one that does not rely on high interest rates.

Feldstein, Martin, American Economic Policy in the 80s,  (Chicago: NBER, University of Chicago Press, 1994.

Galbraith, J.K., Up from monetarism and other wishful thinking, New York Review of Books, August 13, 1981.

Music break 4

This is one of my favorite with Miles. It's awesome how Miles just gets in there right at the onset. Enjoy and stay cool.

PIMCO's Bill Gross: Government must take a leading role in job creation

I was pleased to read this week that Bill Gross, co-founder and managing director of the investment giant PIMCO, believes the US federal government should be doing more to address the unemployment problem. According to Gross, "deficits are important, but their immediate reduction can wait for a stronger economy and lower unemployment. Jobs are today’s and tomorrow’s immediate problem."

Gross's brief also contains some valuable commentary reminiscent of the past insights of the now-retired PIMCO partner Paul McCulley (now with the Global Interdependence Center). Gross's dismissal of the Ricardian equivalence proposition (the notion that budget deficits dampen consumer and investor confidence) and his reference to the ideas of economist Hy Minsky reminded me of this excellent commentary by McCulley.

Also, it's interesting to see Gross giving clear support for more activist fiscal policy measures. In the past, some of Gross's views could have been interpreted as being somewhat dismissive of the effectiveness of Keynesian-type remedies (for instance, see here, p.4).

That being said, if Gross is serious about his plea for more government action, perhaps he should warn the current author of PIMCO's Global Central Bank Focus column, Tony Crescenzi, to ease-off on the anti-inflation rhetoric (see here). First of all, much of what Crescenzi writes in regard to the effects of inflation is not supported by the facts. Secondly, Crescenzi's views on inflation are exactly the type of commentary that opponents of government intervention would use to attack the very proposals now being recommended by Gross.

Increase in household debt slows

Canada's 2011 Q1 National Balance Sheet Accounts were released yesterday. Here are some of the highlights:

• the growth in household net worth as a percentage of income is slowing
• the growth in debt to personal disposable income and household debt to GDP is slowing
• household debt to net worth rises for a second quarter in row after having fallen in mid-2010

The good news is that the household sector is now definitely up to reducing its level of debt. The bad news is that improving household balance sheets will probably take time.

The picture is different with respect to the corporate sector, as the total debt to equity of private non-financial corporations continues to fall to record lows.

On a final point, I've been hearing commentators say that the household sector's position isn't so bad given that the household debt service ratio is still at a reasonable level. That, in my view, is a very simplistic way of looking at things. For instance, the chart below (double click to enlarge) shows that Canada's moderate debt service ratio is essentially being sustained by the low interest rates.* If you look at indicators such as total liabilities or the ratio of debt to net worth, you see that the situation isn't as acceptable as some make it out to be.

To be sure, higher levels of household debt may sustain higher levels of household assets. Also, the fact that debt is greater than income isn't a problem unless household income isn't sufficient to cover for debt payments and living expenses. However, the situation could turn ugly if the value of household assets falls, as this would undermine the basis on which the debt was issued. Therefore, in a context in which some economists are predicting a significant decline in home prices (up to 30 percent in some areas) within the next few years, it would be wise for Canada's monetary and fiscal authorities to tread carefully during this period when households are seeking to reduce their level of debt. Rising interest rates or a decline in household income could trigger financial problems moving forward.

* The chart is based on Statistics Canada data and my own calculations. Data in chart is for persons and unincorporated businesses.

Slowdown in consumer services

Here's a stock paragraph from Statistics Canada's web link to monthly data about the country's services sector:

Our mining, oil and natural gas industries may get all the headlines, but they will not displace service industries any time soon. The services sector dominates the economy. Canadian businesses and consumers rely heavily on the services sector for a wide range of activities.

So what does it mean when the rate of change in consumer services goes negative?

While it's too early to make predictions based solely on this statistic, I think it's important to keep in mind that the federal government has made it clear that it too is in the process of reducing its contribution to the services sector in order to eliminate its budget deficit. Also, if you consider the fact that consumer borrowing is slowing down and export potential isn't what it used to be due to the strong Canadian dollar, it's clear that all there is left to keep GDP growing is business investment, an area that is actually doing well these days.

Perhaps the fall in demand for consumer services is simply a one-off event.* That being said, it should be emphasized that a negative rate of change in consumer services has only occurred once before in the last decade. And that was in 2008 at the start of the last recession.

* Statistics Canada defines consumer service as follows:

Services consumed by households, such as rent (including the rent imputed on owner-occupied housing), transportation, education, medical care, child care, food and accommodation services as well as travel expenditures of Canadians abroad, less travel expenditures of foreigners in Canada. Also includes the current (operating) expenses of associations of individuals and unincorporated businesses.

Corporate vs. government debt: putting things in perspective

From Sally Kohn:

The United States generates approximately $14.5 trillion in GDP each year and carries, currently, $14.3 trillion in debt. That represents a debt-to-income ratio of roughly 1-to-1.

By comparison, here are the debt-to-income ratios of some of the leading corporations in America:

•IBM — 2-1

•Dupont — 3-1

•United Technologies — 3-1

•Boeing — 4-1

•Caterpillar — 14-1

•JP Morgan Chase — 50-1

In other words, IBM borrows twice as much money as it earns annually. Boeing borrows four times more than it earns. And JP Morgan Chase, clearly not too big to borrow, borrows 50 times more than it earns — getting $50 from lenders for every $1 it makes.

I've always thought that reminding people of these facts was a pretty good way of demonstrating how the notion of eliminating public sector deficits is misguided.

And in case you're wondering: no, the analogy isn't faulty. In fact, Bill Vickrey, the late economist and Bank of Sweden Nobel laureate, made a similar case in favor of government budget deficits a decade ago:

If General Motors, AT&T, and individual households had been required to balance their budgets in the manner being applied to the Federal government, there would be no corporate bonds, no mortgages, no bank loans, and many fewer automobiles, telephones, and houses. (p. 2)

Still, some disagree with comparing private and public debt on the grounds that government spending, unlike the product of private industry, is unproductive. If you happen to adhere to this view, I suggest you take a look at "Is Public Expenditure Productive?" by David Aschauer, a former economist of the Federal Reserve Bank of Chicago. According to Aschauer (and contrary to what most people believe), additions to the net capital stock of public infrastructure have a positive impact on overall productivity growth.

Aschauer, D., "Is Public Expenditure Productive?", Journal of Monetary Economics 23, 1989, pp. 177-200.

Vickrey, B., "Fifteen fatal fallacies of financial fundamentalism: A disquisition on demand side economics", Working Paper No.1, January 2000. A copy of the paper is available on the website of the Center for Full Employment and Price Stability, http://www.cfeps.org/pubs/

NY Fed staff: The banking system no longer resembles the fractional reserve banking model

In an article last year, economist Marc Lavoie noted that the financial crisis and the reaction to it have forced monetary authorities to publicly reject some key assumptions of mainstream monetary theory as a way to prevent market participants from misrepresenting and misjudging the effects of recent Fed actions. According to Lavoie, these features of mainstream theory include the notion that increases in bank reserves get "multiplied" into a larger expansion in the broad money supply as banks increase lending, as well as the idea that increases in bank reserves at the central bank result in price inflation.

As these recent blog entry and working paper by NY Fed staff can attest, it's clear that the Fed's public relations campaign to change people's understanding of all this is still underway. No doubt, these efforts are also partly aimed at countering the views recently expressed by certain economists within the Federal Reserve System regarding the potentially inflationary nature of an expanded monetary base.

That being said, I find the title of McAndrews's blog entry misleading. I fail to see how disproving the money multiplier means QE isn't inflationary. In my view, there is a good case to be made that the Fed's asset purchase has ramped up the search for yields, leading banks into speculation resulting in, for instance, commodity inflation. However, the inflationary potential of QE should be viewed as a separate issue given that the resultant inflation would not be a consequence of increased lending caused by the increase in bank reserves.

Here is an interesting excerpt from the working paper by McAndrews et al. explaining how the Fed's authority to pay interest on reserves nullifies the money multiplier process and the notion of fractional reserve banking:

In this note, we present a basic model of the current U.S. banking system, in which interest is paid on bank reserves and there are no binding reserve requirements. We find that, absent any frictions, lending is unaffected by the amount of reserves in the banking system. The key determinant of bank lending is the difference between the return on loans and the opportunity cost of making a loan. We show that this difference does not depend on the quantity of reserves. [...] The current banking system in the United States and worldwide no longer resembles the traditional textbook model of fractional reserve banking. Historically, the quantity of reserves supplied by a central bank determines the amount of bank loans. Through the "money multiplier", banks expand loans to equal the amount of reserves divided by the reserve requirement. However, in many countries, reserve requirements have been reduced either to zero, or to such small levels that they are no longer binding. (p. 1)