Graph of the week: total world debt load

From the WSJ blog:

Source: WSJ

"$223.3 trillion: The total indebtedness of the world, including all parts of the public and private sectors, amounting to 313% of global gross domestic product. 

Advanced economies tend to draw attention for their debt at the government and household levels. But emerging markets are gathering debt at an increasing pace to drive their economic development."

So if the total debt of the world is 313% of global GDP, one has to ask who are we in debt against? Each other? Yes and no. Taking into consideration the dynamic aspect of the story, we are actually in debt against the future generations. This is in fact one of the reasons politicians like to engage in debt spending; instead of increasing taxes on today's voters to fund their current expenditures, they increase the burden against the future generations. This why public debt is often described as a tax against future generations. Someone will eventually have to pay it off.

Notice also how public debt is much larger in developed economies. It is believed that a debt sustainability threshold for developed economies is larger than for emerging economies (in terms of higher interest rate variability and relative riskiness for example). For developed it is believed to be (the allegedly controversial) 90% of debt to GDP, while for emerging markets it is said to be around 60%. So this could be the reason why politicians in developed economies have higher scope for debt financing, under the assumption that it is OK to pile up debt as long as your GDP growth rate is higher than the interest rate (debt payment). Of course this only works to a certain extent. Take the Euro periphery (Greece or Portugal for example). Their GDP growth rates (3-4%) were certainly higher than their debt payments (1%) in the pre-crisis decade, but these low rates were clearly artificial and unsustainable. This didn't stop their governments from overleveraging; quite the opposite it encouraged them to borrow more. 

Akerlof, Blanchard, Romer and Stiglitz (et al) on the state of macro

An impressive group of economists shared thoughts on rethinking macroeconomic policy. Last month in Washington D.C., the IMF has held a conference on the first steps and early lessons on macroeconomic policy with respect to the ongoing crisis. Simply by looking at the title one would say: a quite common topic that has occupied a majority of space and attention in policy-oriented conferences in the last couple of years. Everywhere you turn you will run into some form of a discussion featuring a variety of economists and non-economists (which ones are more interesting?) on the lessons from the crisis, offering their "unique" solutions on how to emerge from it and achieve a robust recovery. However, when such a conference is being held at the IMF headquarters and the speakers are Nobel prize winners or notable economists like Blanchard, Romer, Roubini, Perotti, Tirole, Fischer, Woodford or policymakers like Mervyn King, Andres Borg and John Vickers, then the credibility of the whole thing suddenly increases (I wonder why Rogoff wasn't there?). The four economists from the title were the ones to close the conference with a panel discussion from which they have summarized their comments on VoxEU last week. 

All the speeches are available at the conference webpage (I watched the one on fiscal policy with Borg, Perotti and Roubini). As for the four panelists, here are some of their most interesting thoughts:

George Akerlof draws on some of the findings from the paper by Jorda, Schularick and Taylor (2011): 

"Not only are financial recessions deeper and slower in recovery than in normal recessions, they also have slower recovery the greater is the credit-to-GDP ratio.That is the history. How do their findings reflect on the current crisis? Curiously, it depends upon the measurement of credit outstanding: With bank loans to the private sector as the measure of credit, the US recovery is about 1% of GDP better than mean recovery for financial recessions; When, in addition, the measure of credit also includes credit granted by the shadow banking system, we are about 4% better than the median recovery in financial recessions"

He justifies all the stimuli and bailout policies claiming that they did exactly what they were supposed to - stopped a financial meltdown and led to a recovery which is in his own words much better than how we perceive it: 

"We should have led the public to understand that we should measure success not by the level of the current unemployment rate, but by a benchmark that takes into account the financial vulnerability that had been set in the previous boom"

So Akerlof feels that economics hasn't done anything wrong, calling the macro policies "trial and success", and advocating the same approach for future shocks as well. I couldn't disagree more. 

Olivier Blanchard is much more down to earth:

"Rethinking and reforms are both taking place. But we still do not know the final destination, be it for the redefinition of monetary policy, or the contours of financial regulation, or the role of macroprudential tools. We have a general sense of direction, but we are largely navigating by sight."

He focuses on 6 different policy areas which require new macro tools: financial regulation (in favour of increasing capital ratios), financial sector (whether or not the credit and financial cycles are separated from the business cycle or was it all just a simple AD shock; he also expressed concerns with Woodford's nominal income targeting, and emphasized the crucial role of financial stability), on macroprudential tools (which he rightly points out haven't been particularly successful so far), governance (in coordinating microprudential, macroprudential and monetary policy tools), sustainability of debt (particularly with respect to the Eurozone problems and implications of the ECB), and finally communication (on how credibility of policy-makers can be crucial in certain times). As the typical representative of the "establishment" (as defined by Kling), Blanchard is careful in his conclusions and is committed in acknowledging everyone's views and reaching a sort of a consensus. 

David Romer is more concerned in attempts to avert the next financial crisis. He stresses out evidence that the financial sector is a continued source of shocks (he mentions quite a few examples to verify this) implying that they are more common than we would think, and very hard to predict. This is why Romer is focused entirely on reforming the financial system in order to ease the shock it bears on the real economy; on one side to decrease the risk exposure of financial institutions with the following measures: 

... I am thinking of stronger capital and liquidity requirements, special rules for institutions that create more systemic risk, and restrictions on the form or capabilities of what financial institutions can do, such as ring fencing in the United Kingdom and the Volcker rule in the United States... 

...it is hard to believe that the relatively modest changes along these dimensions ... are really big enough to give us a financial system that is so robust that it is not going to periodically cause severe problems. Shadow financial institutions may escape the rules altogether; rules can be gamed; and shocks can be so large that they overwhelm the moderate changes that were being discussed.... 

There were occasional mentions of very large capital requirements; for example, Allan Meltzer noted that at one time 25% capital was common for banks. Should we be moving to such a system? Amir Sufi and Adair Turner talked about the features of debt contracts that make them inherently prone to instability; should we be working aggressively to promote more indexation of debt contracts, more equity-like contracts, and so on? We can see the costs that the modern financial system has imposed on the real economy; It is not immediately clear that the benefits of the financial innovations of recent decades have been on a scale that warrants those costs. ... The fact that shocks emanating from the financial system sometimes impose large costs on the rest of the economy implies that there are negative externalities to some types of financial activities or financial structures, which suggests the possibility of Pigovian taxes.

...and on the other side to make the real economy more resilient to frequent financial shocks (with some standard measures but also calling for fiscal rules and constraints). Altogether Romer asked a lot of questions and made a lot of decent implications for deeper thinking on the issue of preventing future shocks. It's not that all of his suggestions would work (in my opinion), but he is correct in that the standard solutions to these issues need to be reexamined. 

And last but not least, Joe Stiglitz focuses on reforming faulty, inefficient economic models. He too notes that financial instabilities and crises are a common occurrence. 

In a very fundamental sense, the crisis is still not fully resolved – and there’s no good economic theory that explains why that should be the case. Some of this has to do with the issue of the slow pace of deleveraging. But even as the economy deleverages, there is every reason to believe that it will not return to full employment. We are not likely to return to the pre-crisis household savings rate of zero – nor would it be a good thing if we did. Even if manufacturing has a slight recovery, most of the jobs that have been lost in that sector will not be regained. ... 

Economies that have had severe financial crises typically recover slowly. But the fact that things have often gone badly in the aftermath of a financial crisis doesn’t mean they must go badly. This is more than just a balance sheet crisis. There is a deeper cause: The United States and Europe are going through a structural transformation. There is a structural transformation associated with the move from manufacturing to a service sector economy. Additionally, changing comparative advantages requires massive adjustments in the structure of the North Atlantic countries.

What Stiglitz describes here is a typical structural shock that hit our economies. I agree with his assessment completely (some of my previous texts on this issue: see here, here or here). He correctly emphasizes that reforms currently undertaken were insufficient and half-way measures (e.g. the too-big-to-fail problem is now even worse). The set of reforms he calls upon concern (1) the provision of credit, (2) stability (lower exposure to risk - similar to Romer's suggestions), (3) distribution (of income), (4) stronger government correction of market failures, and (5) quantitative not price interventions. Stiglitz unfortunately fails to mention any of the institutional reforms that are necessary to adapt to the technological shock and the change in compartivie advantages that he recognizes. All the proposals he stresses out are just an adjustment in my opinion, they don't offer the necessary and much needed change of the growth model.

Altogether, it is always enjoyable to encounter a wide variety of different and prudent ideas on macroeconomic policy. I look forward to more attempts such as these. Perhaps by over-thinking about some of the issues we may reach a new equilibrium of thought? Just perhaps... 

Why do people in societies with low levels of trust advocate higher regulation?

An insightful working paper by Aghion, Algan, Cahuc and Shleifer can give us a possible answer to this question:

Source: Aghion, Algan, Cahuc & Shleifer (2009): "Regulation and Distrust"
NBER Working Paper 14648, pg. 9

"In a cross-section of countries, government regulation is strongly negatively correlated with social capital. We document this correlation, and present a model explaining it. In the model, distrust creates public demand for regulation, while regulation in turn discourages social capital accumulation, leading to multiple equilibria. A key implication of the model is that individuals in low trust countries want more government intervention even though the government is corrupt. We test this and other implications of the model using country- and individual-level data on social capital and beliefs about government’s role, as well as on changes in beliefs and in trust during the transition from socialism."

The emphasized sentence is particularly important from a policy-making perspective in transitional economies, which are more likely to have lower levels of interpersonal trust, in addition to a low level of trust in corporations, the legal system or political institutions. The essential finding is that in countries with lower levels of social capital, people support more government intervention in the economy. This becomes even more obvious in times of crises. And what is particularly intriguing is that even though the people know the government is corrupt and inefficient they would still prefer things to be (over)regulated rather than to be left "at the mercy" of an uncivic surrounding. Or the generalize, "people in countries with bad governments want more government regulation". 

Formal vs informal institutions 

This seemingly paradoxical finding is somewhat logical if one thinks about the interrelation of a country's formal and informal institutions. If formal institutions are weak and inefficient, meaning that if the rules in society are poorly defined and offer too much scope for corruption, then informal institutions will adjust accordingly and this will first and foremost affect the accumulation of social capital (people start loosing trust in the system). The relationship works in both ways; if the 'rules of the game' are defined to be inclusive, then the accumulation of (let's call it) democratic capital will eventually lead to a change in customs, norms and trust in a society (i.e. our informal institutions).

The intuition of the paper explains this further:

"In this model, when people expect to live in a civic community, they expect low levels of regulation and corruption, and so invest in social capital. Their beliefs are justified, and investment leads to civicness, low regulation, and high levels of entrepreneurial activity. When in contrast people expect to live in an uncivic community, they expect high levels of regulation and corruption, and do not invest in social capital. Their beliefs again are justified, as lack of investment leads to uncivicness, high regulation, high corruption, and low levels of entrepreneurial activity. The model has two equilibria: a good one with a large share of civic individuals and no regulation, and a bad one, where a large share of uncivic individuals support heavy regulation."

Measuring social capital

Which of the two equilibria does your country belong to? Perhaps this will help:

Source: Interpersonal trust index

Those with a darker red color seem to have less interpersonal trust (the overall level of the trust index is calculated as 100 + (% of those who say that most people can be trusted) - (% of those who say most people cannot be trusted)). See here for more details, individual country data, continental maps and tables. Some of the countries on the bottom of the list are Trinidad & Tobago, Turkey (!), Rwanda, Cambodia, Indonesia, Ghana, Brazil, and from European countries Cyprus and Portugal. On the other hand, the top 10 are the 'usual suspects' Norway, Sweden, Denmark, Finland, Switzerland, New Zealand, Australia, Nederlands, Canada (countries that top the indicators of economic and personal freedoms), however with a surprising inclusion of China (4th place), Saudi Arabia (7th) and Vietnam (8th). A possible explanation of the lower end of the spectrum could be (among others) conflicts in the past or large ethno-linguistic fractionalization. On the upper end the three outliers could perhaps be explained by lower inequality or even a long-lasting autocratic rule.

There is an interesting line to draw here between the countries of the former socialist bloc and countries who haven't interrupted their socialist/autocratic rule. Almost all countries that have undergone a transition from socialism (even those with an early EU accession) experienced lower levels of interpersonal trust.

On the other hand the 3 outliers (we can very well add 13th ranked Belarus to the picture) have had a long time without a political crisis. This has perhaps led to lower uncertainty over regime change, and despite the regime being autocratic, the population has been able to adapt to the existing environment and accumulate a higher level of social capital. Countries which have experienced significant political change and/or persistent conflicts have naturally depleted their accumulated social capital.

Transition from socialism and depletion of social capital 

It would be very interesting if we could compare the relative levels of trust a couple of years before the fall of socialism and a couple of years after. Going back to the Aghion et al paper, their finding is that after the transition social capital in every transitional economy decreased.

In the paper the authors interpret a transition from socialism as a radical reduction of government control in societies with low levels of trust. Their consequential prediction is that this has increased corruption and the demand for government control, while decreasing output and short-run social capital. But what if these countries didn't have a low level of trust before the transition? What if these countries were similar to Vietnam, China, Saudi Arabia or Belarus? That would imply that the transition itself (perhaps because it was done under suspicious conditions) led to a depletion of the existing social capital.

The paper observes four different datasets in which the authors asses trust (from the World Values Surveys): 1981-84, 1990-93, 1995, 1999-03 (so exactly the necessary series to observe the effect of the fall of socialism). They use this dataset for 57 countries (most of the aforementioned included).

The static correlation data imply the following relationships between trust in different types of institutions within a country (particularly interesting is the regulation on the minimum wage - 3rd graph): 

Source: Aghion, Algan, Cahuc & Shleifer (2009): "Regulation and Distrust"
NBER Working Paper 14648

However, more importantly they prove the relationship between low trust and the demand for more regulation (the demand for trust variable is calculated from a couple of World level surveys according to the specific questions asked - see the paper for details). 

In their further analysis they imply similar findings; positive relationship between distrust and more government intervention (measured through both price and wage controls for example), but I failed to see a coherent analysis linking the levels of distrust a couple of years before the start of the transition and after the transition. They only look at the level of distrust in transitional economies in the year 1990, and come up with the following graph:

Source: Aghion, Algan, Cahuc & Shleifer (2009): "Regulation and Distrust"
NBER Working Paper 14648

Along with this table in which they show the initial higher distrust in transitional economies with comparison to OECD countries.

I would argue that this is still a bit late to conclude that former socialist countries had initial lower levels of trust. The transition varied for different countries, but in 1990 it was evident to the people that things were changing and uncertainty was surely high, regardless if the people were looking forward to the change or against it. I would rather compare the level of trust in 1981-84 to get a more precise picture.

Evolution of social capital 

The paper actually does look at the evolution of social capital in transitional economies and according to the graph below they conclude that distrust has increased in all transitional economies.

Source: Aghion, Algan, Cahuc & Shleifer (2009): "Regulation and Distrust"
NBER Working Paper 14648, pg. 9

Their statistical estimates verify this trend. However, they tend to draw a conclusion from this which I find rather far-fetched: "Liberalization of entrepreneurial activity starting from a low level of social capital has increased corruption, invited a demand for greater state control of economic activity, and reduced trust." How can we be so sure that they started of with a low level of social capital? The authors use the figure depicting the level of trust in 1990, but this, in my opinion, is not enough to draw such an implication. 

A noteworthy finding 

Overall, the conclusions of the paper are certainly noteworthy in helping us understand and explain some of the interaction between formal and informal institutions and how they help shape economic outcomes. This paper has certainly proven that social capital (regardless of how we think about it, define it or measure it) affects the society's institutional development, and more importantly the people's preferences towards government intervention, despite its (in)efficiency. 

I would, however, like to see an extension to the paper trying to explain the aforementioned autocratic outliers. If they undergo a transition to an inclusive democracy, will it be easier for them to do so because of high existing levels of social trust, or will they succumb to the same scenario of transitional economies? The Aghion et al paper indirectly implies that their high levels of social capital will be helpful, but I fear that when regime uncertainty kicks in, social trust immediately declines. This wouldn't mean that social capital isn't important in shaping formal institutions, it simply means that the initial line of causality goes the other way around. 

Inclusiveness vs extractiveness, not democracy vs autocracy

One of the quintessential questions in political economy is which is better for growth: democracy or dictatorship (with the proper preassumptions of course)?

Here's a graph from the Free Exchange blog (click to enlarge):

These are the implications:

"Does economic growth go hand-in-hand with democratic regimes? Not necessarily: correlation does not imply causation. One group of economists found growth induced democracy in East Asia; democracy did not lead to growth. They compared North and South Korea, which were both poor in 1950 and under dictatorial regimes from the end of the Korean War until 1980. From 1980, per capita incomes diverged. The same year South Korea began democratising. But South Korea’s better institutions developed due to dictators’ policy choices, they say.

Others, including Daron Acemoglu and James Robinson, attribute this type of growth to political decision-making. “Extractive institutions” sometimes develop as elites feel more secure and seek their own ends, they say. “Such growth takes place when elites find it in their interest to allow new technologies and institutional changes necessary for economic growth.”

Paul Collier has controversially argued that authoritarianism can be good for growth. He would also say South Korean growth was successful due to its homogenous society. Its foreign immigrant population only reached 1m in 2007, and the majority are Chinese. In ethnically diverse societies only democracy can work for growth, says Mr Collier, because autocratic leaders with a narrow support base are otherwise tempted to siphon off national income. That explains why diverse India, with three major ethnic groups, four key religions, and 15 official languages, had no choice other than democracy-led growth."

But the real issue is not whether a democracy will give us a higher technological breakthrough, but that an inclusive democracy will result in the same. There is a big difference between sustaining the efficiency of institutions in different types of democracies. A democracy in itself does not imply efficient institutions, according to Mancur Olson (2000). Only a fully inclusive democracy will foster institutional efficiency. This implies it has both inclusive political and economic institutions such as the proper functionality of the legal system (rule of law, enforcement of contracts), protection of property from expropriation by a number of elites (financial, corporate or political), and finally an egalitarian system which offers equal opportunity, social mobility and basic human rights (freedom of speech, freedom of assembly, etc.). (A hard number of conditions to fill out, but some countries have managed to do it, right?) 

If we add some "bad" democracies in the the graph above, I'm sure they would turn out as outliers rested in the lower or perhaps lower middle technological quadrant and a high-income quadrant. Even though the graph itself is very perceptive as it tell us exactly of 

As for Paul Collier's South Korean argument, isn't it true that the same homogenous society exists in North Korea as well? Prior to the separation of the Korean peninsula, wasn't this a homogenous society, living on the same geographical position, having the same climate, same historical context, same cultural origin? And yet they diverged into two extremes 60 years later. Isn't the same story applicable to East and West Germany? Or any of the examples mentioned in this post? The reason for such divergence can only be explained by the quality of institutional formation. And one may argue that Korea had a state-led growth that resulted in economic inclusiveness  but it is also a fact that at one point the people in Korea demanded more political freedoms. Just like in 19th century England, where the rising wealth of merchants and manufacturers forced enfranchisement from the aristocratic elites (who had won their power in a similar way from the monarchs 200 years earlier). The South Korean example is more likely to be one of a gradual democratic consolidation. 

The Chinese outlier? 

So how does one explain China? Is it an outlier in the theory of inclusive institutions? Not likely. In previous blog posts I have made certain inferences on why the Chinese growth model was successful, and how a huge level of competition drives China forward (the most notorious example is the Foxconn iPhone factory, where there is a barbed wire around the factory, not to prevent the workers from "escaping" but rather to prevent others from coming in - that's what I call Social Darwinism). However, just like Soviet Russia and a multitude of different examples of states with extractive institutions that were able to achieve initial substantial growth, one cannot grow for an infinite amount of periods based on pure capital accumulation. The basic Solow-Swan growth model teaches us this: at one point more capital will fail to increase productivity and will fail to result in more economic growth. A technological breakthrough will be necessary to break the gridlock and move the economy towards a new steady-state equilibrium. The emphasis is therefore on persistent innovation and new technologies. These are the main long-run factors of economic growth in the West. Of course, they are supported by an inclusive democratic system and institutions that favour free enterprise.

In China, things are a bit different. Instead of being an innovation nation, China is an adapter of technologies. So was Japan back in the 50-ies, but they managed to switch to an innovation nation eventually. Can China do the same? Certain factors point out otherwise. There is a particular example pointing out to the Chinese anti-entreprenuership climate - the example of a business owner thrown in jail for competing against the state-owned companies and taking larger portions of their markets. With such protection of inefficient industries, once the economy runs out of steam it will be obvious that there is nothing to replace the previous growth model and a political crisis will be looming. There is no telling on what the final outcome will be; will China switch to more inclusive political instituions, or will the communist party apply stronger methods of repression? There has already been a hint towards changes in China, with the new PM announcing plans to fight corruption and curb government power, but it is left to be seen whether or not these threats are credible enough.

The reverse savings glut

The personal savings rate in the US has experienced a steady decline in the past 30 years, up until the recent crisis:

Source: FRED

From the Economist:

"The drop began in the 1980s, perhaps because the Great Moderation made people less fearful of economic uncertainty. When uncertainty returned during the financial crisis, and as credit conditions tightened, the saving rate shot up. Wage stagnation may also play a role; people expected better living standards and cut back on saving to raise consumption. The saving rate fell again this past January. That may reflect greater economic optimism or the return of the full payroll tax, which lowered take-home pay. Rather than decrease consumption people may have saved less."

A declining savings rate is a worrying signal to the economy as it increases its vulnerability to crises. The reasoning is clear; the higher the leverage of the private sector in pre-crisis times, the stronger the deleveraging will be after the shock. Private sector savings will go up and borrowing will be scarce, implying a lot of underutilized resources.

Why savings went down? 

There are several plausible explanations for the decline of the savings rate and an increase of borrowing, particularly in the pre-crisis decade (see graph below - red line). Some claims have been made that households, rather than keeping money in liquid assets, were increasing savings in the form of retirement accounts. The Wall Street Journal had a good report on how automatic enrollment in 401(k) accounts has boosted the number of savers but has decreased the average savings rate. The popularity of such pension accounts could also be accounted to an ageing US population. Another interesting factor in this story is that recently, there has been an increase in the number of savers drawing on their pension plans too early (and hence accepting a penalty for early withdrawal). This only further stresses the gravity of the current financial situation among US households. 

The reasons behind the relative decline of savings in liquid assets can be either slow growth of average income in the past 30 years, or a low return on low-risk savings discouraging people to save. The New Yorker did an interesting story on the myth that easy money is hurting savers, recognizing the real problem in which there are a lot of people out of jobs and out of money, so basically cannot save anything at all. Taking all these factors combined (slow personal income growth, low returns on equity, a switch to non-liquid retirement accounts), they could offer a potentially reasonable explanation of why people borrowed more in the past 30 years and saved less, thus exposing themselves to more financial risk. This story is similar to Rajan's Faulty Lines argument.  

The reversal of the trend (following the housing bubble)

So with household debt increasing during the Great Moderation (due to a variety of reasons), the current deleveraging process can hardly come as a surprise.

Source: FRED

However, total household debt as percent of disposable income (green line) didn't experience significant increases until the last decade, and is now on the same level as it has been in the beginning of the 1980s. The housing bubble and the consequential crisis acted as an immediate reverse signal to households who started to save and deleverage, and this process is still far from finished. The balance sheet recession argument seems to be in order; after a significant shock personal savings go up, borrowing goes down, and there is a consequential lack of liquidity in the system. As for the argument for more liquidity to be necessary provided by the government, in the graph below we can see that this has indeed happened during the current shock. The government has provided excess liquidity and increased spending, but this has failed to prevent the deleveraging process, while the savings rate remained high.

Source: FRED

One could argue that the deleveraging process would have been even stronger without government involvement, but the effect on savings would have still been more or less the same. This is probably due to the phenomenon of a reversed relationship between public and private savings.

So how do we reignite borrowing and reverse the deleveraging process of households and businesses in order to start growing? Who is to say that this is the right approach anyway? A lot of risk has been accumulated (it was perceived to be non-existent) in the pre-crisis decade and as a consequence the system has to undergo significant deleveraging (on the corporate side). The problem of inequality and low income growth on the household side is hardly a short-run issue that can be solved by temporary liquidity.

The proper incentive to break this process is not to cramp the system with public sector spending, but rather to offer different incentives to households and businesses to restore their confidence. The psychological effect of governments restoring confidence doesn't seem to work if these governments are constantly being threatened by unsustainable public finances and outdated welfare models. This is why the confidence to the private sector needs to be restored via cost-oriented incentives. There is no better incentive for a business than a cost-cutting one. It frees up funds and creates the scope for higher production and eventually higher employment.