The financial cycle*

“Macroeconomics without the financial cycle is like Hamlet without the prince,” says Claudio Borio, Head of the Economic and Monetary Department of the Bank for International Settlements in Basel. Borio has long argued that the business cycle alone can’t explain some features of expansions and recessions. The Great Depression and the prolonged decline of Japan in the 2000s were unusual because financial factors were at play. Likewise, to understand the poor growth of advanced economies since 2007 one must look at the interaction between asset prices, debt, and the real economy. Borio and the BIS have lead this research agenda, tracing the links between financial and business cycles.

The “financial cycle” isn’t a new idea, of course. Hyman Minsky and Arthur Kindleberger became well-known for their work on credit cycles and market manias. Michael Bordo, Guillermo Calvo, Gary Gorton, and Carmen Reinhart, to name just a few, all had done important research on financial vulnerabilities by the early 2000s. Franklin Allen and Douglas Gale did the work for their book “Understanding financial crises” well before 2007.

But, for the best part of the 20th century, mainstream macro models didn’t integrate the financial sector, as Calvo (2013) recounts beautifully. Financial capital in those models was allocated, without a hiccup, in aseptic markets. Most models ignored banks altogether. Only after the earthquake of 2008-2009 have economists rediscovered the role of financial factors in macroeconomics. Today the economics department of every major bank tracks macro-financial risks.

BIS updates are required reading if you want to keep an eye on the global financial cycle. This year’s Annual Report includes a riveting chapter on it, based on Borio’s research, which forms the basis for this column.

What is the financial cycle?

What is this “financial cycle” anyways? There’s no set definition, but for Borio it stands for medium-term fluctuations in financial variables. Those fluctuations come with “self-reinforcing interactions between perceptions of value and risk, attitudes toward risk, and financing constraints, which translate into booms followed by busts” (Borio (2012)).

In plainer English: Rapid increases in credit lift property prices, which in turn increase the value of collateral and thus potential credit. The rising supply of credit relaxes financing constraints, whereas higher property prices change how investors perceive value and risk. Those interactions aggravate swings of real variables (output, profits, employment, etc.), and distort how capital spending is allocated across industries. The cycle usually ends in financial distress or even a crisis. Asset values plunge, and borrowers reduce their leverage, until the expected return on capital rises enough to lure investors back in.

Five features of the financial cycle stand out. First, it’s much slower than the business cycle. Two or three recessions typically go by, over 15 to 20 years, before a full financial cycle is through. Drehmann, Borio, and Katsaronis (2012), studying seven mature economies between 1960 and 2011, find that the average cycle lasted 16 years. The United Kingdom, for instance, experienced only three financial peaks in that period (in 1973, 1991, and 2009).

The second feature is that peaks in the financial cycle are typically followed by systemic financial stress. After many years of rising asset prices and growing debt, asset values are out of kilter with fundamentals. If a shock then makes credit contract, or leads investors to reassess value and risk, the value of collateral plummets and borrowers can’t service their debt. Severe distress or an outright banking crisis follows.

The third feature —and a direct result of the second— is that recessions are deeper when they coincide with the contraction phase of the financial cycle. Drehmann, Borio, and Katsaronis (2012) find that, on average, gross domestic product drops 50% more than in regular recessions (-3.4% versus -2.2%).

Fourth, the length and amplitude of the financial cycle depend on background conditions. For instance, a lightly regulated financial industry eases financial constraints, which speeds up the loop between investors’ views on value and risk, risk tolerance, and funding conditions. Similarly, if monetary authorities put the emphasis on reigning short-term inflation, interest rates will be too low when financial booms develop in a low-inflation environment. As another example, globalization may have made potential output seem higher, and inflation lower, than their sustainable levels, stoking credit growth and asset prices.

Fifth, and last, financial cycles are often in sync. That’s because common factors drive cycles across countries. Financial capital moves across borders, equalizing risk premia and funding conditions. As a result, financial crises happen in groups. Lowering barriers to foreign investment has only intensified this clustering.

Where are we now?

Where are we in the financial cycle? To trace its path, Borio and his colleagues use the medium-term co-movement of three things: credit to the non-financial sector, the credit-to-GDP ratio, and real housing prices. The medium-term component is extracted from each individual series using a technique called bandpass filtering, which removes fluctuations with a frequency shorter than 8 years, or longer than 30. Then a simple average rolls the three filtered series into one.

Using data through 2013, the BIS Annual Report finds that countries are at different stages (see Figure 1). Southern Europe is in the downswing phase of the cycle. In that region –Spain, Portugal, Italy, and Greece— the global recession was the preamble to the 2010 sovereign debt crisis, when the cycle peaked (a little earlier in Greece). In 2014 I see that real property prices and credit are still falling, but at decreasing rates.

Figure 1

Source: Figure IV.I from the 84th Annual BIS report. Data (xlsx).

Credit and property prices in the United States and the United Kingdom bottomed out recently, and are now clearly recovering. Japan is on the rising phase too, but its current cycle began in 2005. Much further ahead in the expansion are Canada and Australia, where the financial cycle merely took a pause in 2007-2009. Switzerland is in a “boom.”

Clearly booming as well are some emerging markets (EMs): China, Brazil, Turkey, and Asia-Pacific. (Asia-Pacific here includes: Hong Kong, Indonesia, Malaysia, Philippines, Singapore, and Thailand. The report doesn’t include Latin American countries, other than Brazil.) In those countries the current financial cycle started in the early 2000s, after the Asian financial crisis. The cycle continues today, briefly interrupted by the fallout from the 2008-2009 crisis.

In the U.S., U.K., and other countries where the cycle peaked in 2007, monetary easing cushioned the fall. In countries , on the other hand, that didn’t have a financial crisis, the surge of global liquidity has catapulted borrowing and housing values. And, almost everywhere, low interest rates have inflated equity prices and raised risk tolerance. Monetary stimulus is not a free lunch.

What might kill those booms? Rising interest rates are a prime concern, as seen in the summer of 2013. India, Indonesia, Brazil, and a few other EMs had a mini-crisis on the mere possibility the Federal Reserve would begin tapering asset purchases. In 2015, both the Fed and the Bank of England will likely raise policy rates, although they will loudly telegraph the move well before it happens.

Another worry is a sharp fall of income. That would make debt more difficult to service. Commodity exporters nearing boom-like conditions—Australia, Canada, Norway, Latin America—would be at risk of a financial crisis if China’s demand slowed down quickly.

One more source of fragility, as the BIS report says, is the clout of asset managers. Large investment funds account for a significant share of all EM assets. Even a small reallocation of their portfolios can trigger large price swings. The rise of indexed products makes this matter worse. Also, the shift towards exchange-traded funds (ETF) may have made the market more volatile, as more and more investors view EM ETFs as liquid, short-term positions (Figure 2).

Figure 2

Source: Figure IV.6 from the 84th Annual BIS report. Data (xlsx).

Timing the market is just as difficult as forecasting the peak of the financial cycle. Early warning signals exist, but no dashboard is infallible. Most predictions of financial downturns come months or years too early. Claudio Borio himself warned of financial vulnerabilities in 2003, four years too soon. When the wolf fails to show up, the sheep keep on partying. Next time won’t be different.

References

Allen, Franklin, and Douglas Gale (2007), “Understanding financial crises,” Clarendon Lectures in Finance, Oxford University Press.

Bank for International Settlements (2014), “84th BIS Annual Report, 2013/14,” Bank for International Settlements.

Borio, Claudio (2012), “The financial cycle and macroeconomics: What have learned?,” BIS Working Papers, 395.

Calvo, Guillermo (2013), “Puzzling over the anatomy of crisis: Liquidity and the veil of finance,” Columbia University, mimeo.

Drehmann, Mathias, Claudio Borio, and Kostas Katsaronis (2012), “Characterizing the financial cycle: Don’t lose sight of the medium term!,” BIS Working Paper #380.

*This article was prepared for the October/November issue of Morningstar magazine, my employer's publication.

What is economic overheating? An introduction

In a few of my recent posts I have referred to “economic overheating.”

I don’t think I ever encountered the term “economic overheating” in any of my undergrad or graduate economics classes, as an economist colleague reminded me recently. I don’t think I ever found the term in any seminar or working paper, or during formal or informal conversations with fellow students or professors. Perhaps at the schools I attended "overheating" is like "Voldemort": a word that it is best to avoid. At any rate, you won’t come across the word easily in academic journals and textbooks. But I do find “overheating” in the media, and in policy and financial reports in the non-academic sphere. Even the IMF (see the foreword, to begin with) and some people at the Federal Reserve use it.

So, for those people who do use the term, what does “overheating” mean? After pondering this for a few weeks, I realize that I can characterize the term, but not define it. Here’s my humble characterization:

Economic overheating is a syndrome characterized by a majority of the following symptoms: increasing growth of credit to the private sector; rising growth of asset prices, including real property prices and equity prices; a decline of the current account balance (declining surplus or widening deficit); a decline in the national saving rate; above-trend growth of private investment or consumption; lower-than-average (and possibly negative) real interest rates; increasing external net financial inflows; above-trend currency appreciation; a decline in foreign exchange reserves; and above-trend output growth and below-average unemployment.

Before you spring off your chair and start firing off a reply, let me acknowledge the lack of objectivity or even preciseness of this characterization. What is exactly “a majority” of these symptoms? More than 50%? Can we take a weighted average? It is difficult to say, but the more symptoms I observe, the more confidence I would have in my diagnostic of overheating.

Also, it seems like I cannot decide whether we should observe a rising growth rate (the second derivative), a deviation from trend, or a deviation from the long-term level. I think it depends. Depending on the variable and the economy under consideration, one should consider deviations of the level from the trend level, deviations from a long-term “equilibrium” level, or deviations of the growth rate.

Some commentators throw in increasing inflation as a typical sign of overheating, but that seems to require the Phillips curve, which I am not ready to assume.

I must insist that not every episode of overheating manifests itself in all the symptoms that I listed above, and my list may not be exhaustive. Also, notice that there may be overlap among these manifestations. I would also like to add to the list a symptom from the banking sector, but at this point I cannot characterize it (would it be a rising ratio of loans to deposits, or a declining capital ratio, or something else?).

Implicit in the concept of overheating is the idea that the economy is developing “imbalances,” both internal and external. Certain parts of the economy are growing “too fast for their own good,” and these imbalances are often accompanied by leveraging. Some analysts stress the real economy side of overheating (unemployment, output growth), while others put the emphasis on the financial side (credit, asset prices).

OK, so we have a list of manifestations of economic overheating. But what are the “deep” economic forces that produce it? The concept of overheating is begging for an economic model. I find “Keynesian” or “neo-Keynesian” too vague of a description. In fact, if you remove unemployment, output and inflation from the list of symptoms, the model might have very little “Keynesianism” left in it. Narrowing down the features of such model is beyond the scope of this blog post, but I do hope to revisit the topic and refine the definition and characterization of this thing they call “economic overheating.”

I encourage readers to chime in, particularly for this post, in the comments.

Links 20130614: TV, Indonesia, Philippines

1. Most European countries have a public television network, at least partially funded by the government budget. Greece just decided to turn its TV off, at least for a while. This brings back to mind a question: Do countries need a public TV? What kind of TV should that be?

To the first question, one could argue that certain types of broadcasting are a public good, in the economic sense of the word. I am more and more skeptical about this argument, as the cost of producing and distributing media content has declined dramatically in the advent of the internet. Also, my experience watching Televisión Española (TVE, the public network in Spain) is that the average quality and breadth of their programming is not significantly higher than that of its private sector competitors. If anything, the liberalization of the sector during the 1980s and 1990s led, I would argue, to a progressive deterioration in the quality of TVE's programs. Rather than forcing competitors to increase their quality, TVE was dragged down into a race for ever dumber, mass-pleasing programs. Another argument against the public good view is that the objectivity of the content produced by public TV is greatly undermined by the very fact that it is funded by the government, thus diminishing the social welfare produced by this supposedly public good.

If, nonetheless, you must have a public TV network, what should it look like? If the public TV must cater to the median voter, but the content that is appealing to that voter is already being provided by private sector networks, public TV is a waste of taxpayer money. I would advocate for keeping it as small as possible: news, weather, parliament debates and announcements,... that sort of thing.

2. Indonesia's terrible school system.

Education experts say less than half of the country's teachers possess even the minimum qualifications to teach properly and teacher absenteeism hovers at around 20 percent. Many teachers in the public school system work outside of the classroom to improve their incomes.

Corruption is also rife within schools and universities - with parents often having to pay bribes for their children to pass examinations or pay for services that should be provided by the state.

Indonesian Corruption Watch claims there are very few schools in the country that are clean of graft, bribery or embezzlement - with 40 percent of their budget siphoned off before it reaches the classroom.

Meanwhile, millions of dollars in foreign aid is poured into the country's education system despite the government spending only a very small proportion of its GDP on schooling. And some international observers are asking why Indonesia still relies on external funding for school construction given that it has been listed as a middle income country by the World Bank.

And here is the Pearson ranking of countries by cognitive skills and education attainment, where Indonesia ranked last in a list of 40 countries. Pearson has posted on its Learning Curve project lots of fun graphic tools and data that allow you to compare countries along many dimensions of human capital.

3. Is the Philippines overheating? You know there is something wrong when a country is growing at 7.8% (more than any of its neighbors), its unemployment rate is in its bottom quintile within the last 20 years, and yet the central bank governor feels compelled to go out and reassure the markets by saying that they have more scope to cut interest rates to "further support growth." That is what is happening now in the Philippines. And the problem is, I think, that the country is experiencing a classic case of excessive capital inflows and shrinking current account balance, currency appreciation, plus a stock market bubble and possibly a property bubble.