Financial Networks and Contagion

"Financial Networks and Contagion," a recent paper by Matthew Elliott, Benjamin Golub, and Matthew Jackson, uses network theory to study how financial interdependencies among governments, central banks, investment banks, and other institutions can lead to cascading defaults and failures.

Source: Elliott et al. 2013

While the model is quite technical, the main theoretical findings are fairly intuitive. They define two key concepts, integration and diversification. Integration refers to the level of exposure of institutions to each other through cross-holdings. Diversification refers to how spread-out the cross-holdings are; in other words, whether a typical organization is held by many others or just a few. The key finding is that at very low or very high levels of integration and diversification there is lower risk of far-reaching cascades of financial failures. The risk of a far-reaching cascade is highest at intermediate levels of integration and diversification. The authors explain:

"If there is no integration then clearly there cannot be any contagion. As integration increases, the exposure of organizations to each other increases and so contagions become possible. Thus, on a basic level increasing integration leads to increased exposure which tends to increase the probability and extent of contagions. The countervailing effect here is that an organization's dependence on its own primitive assets decreases as it becomes integrated. Thus, although integration can increase the likelihood of a cascade once an initial failure occurs, it can also decrease the likelihood of that first failure... 

With low levels of diversification, organizations can be very sensitive to particular others, but the network of interdependencies is disconnected and overall cascades are limited in extent. As diversification increases, a "sweet spot" is hit where organizations have enough of their cross-holdings concentrated in particular other organizations so that a cascade can occur, and yet the network of cross-holdings is connected enough for the contagion to be far-reaching. Finally, as diversification is further increased, organizations' portfolios are sufficiently diversified so that they become insensitive to any particular organization's failure."

Near the end of the paper, they illustrate the model using cross-holdings of debt among six European countries. The figure above is their representation of financial interdependencies in Europe. They conduct something akin to stress tests, simulating cascades of failures under various scenarios that very roughly approximate conditions in 2008. The simulations find that, following a first failure in Greece, Portugal is fails from contagion. After Portugal fails, Spain fails due to its large exposure to Portugal. The high exposure of France and Germany to Spain causes them to fail next in most simulations. Italy is always last to fail due to its low exposure to others' debt. They emphasize that this is intended only as an illustrative exercise at this stage, but could eventually be refined and incorporated into analysis of failure and contagion risk.

*Edited to fix my mistake pointed out by Phil.

Possible Futures for the European Banking Union

In September 2012, the European Commission proposed a single supervisory mechanism (SSM) for banks as a preliminary step towards a European banking union. Plans for the union continue this week, as European Union leaders meet in Brussels to set new rules regarding how future bank bailouts will be paid for.

 Earlier this year, I attended the Future of the Euro Conference at UC Berkeley. One of the conference panels was about banking unions. The panelists all draw upon economic history to discuss the possible future of a European banking union. Here are videos of their talks:   

The OMT Goes to Court

Imagine a scheme that would transfer money from Spain, Greece, and Italy to Germany. According to Paul De Grauwe and Yuemei Ji, this is precisely what the European Central Bank's Outright Monetary Transactions (OMT) program has the potential to do. Yet 35,000 Germans have filed complaint against the scheme.

The Federal Constitutional Court of Germany is undergoing deliberations on the legality of the European Central Bank's OMT program. The program was announced last August, when the ECB announced that it would be willing, in certain scenarios, to buy government bonds without limit. Even though the ECB has not yet purchased government bonds under this scheme, the announcement alone reduced problematically-high bond yields in Italy and Spain, leading Mario Draghi to declare the program a big success.

Andreas Vosskuhle, president of Germany's Federal Constitutional Court, however, said Tuesday that the court will not take the success of the OMT into account when deliberating its legality or constitutionality. Andreas Wiedemann has written an excellent summary of the hearings on June 11 and 12. The court is expected to make a ruling some time after the German elections on September 22.  Mark Thoma links to a note by Helmut Siekmann and Volcker Wieland on the legal issues of the case. They begin by explaining several types of criticism of the OMT.

The first criticism is that "the ECB has ventured too far into the terrain of fiscal policy by announcing such potentially unlimited government bond purchases. The announcement itself is likely to cause delays in the implementation of necessary fiscal and structural adjustments by national governments, because it has reduced market pressures via government financing conditions." Moreover, the "strict and effective conditionality" prerequisites for OMT measures "can also be interpreted to indicate that the envisaged measures fall outside the area of monetary policy as these conditions serve to achieve other economic and fiscal objectives." An additional point of contention is that "critics fear that the independence of the ECSB and of the members of their decisionmaking bodies is jeopardized by the large-scale transfer of credit risks from the private and public sector to the ECSB."

In "Fiscal implications of the ECB’s bond-buying programme," De Grauwe and Ji argue that the fears that German taxpayers may have to cover losses made by the ECB are misplaced--based on a misunderstanding of solvency issues facing central banks. They say, in fact, that German taxpayers are the main beneficiaries of such a bond-buying program. They begin by explaining that a private company is said to be solvent when its losses do not exceed the value of its equity, which is the expected present value of future profits, but the same is not true for a central bank.

"A central bank can issue any amount of money that will allow it to 'repay its creditors', i.e. the money holders... Contrary to private companies, the liabilities of the central bank do not constitute a claim on the assets of the central bank. The latter was the case during gold standard when the central bank promised to convert its liabilities into gold at a fixed price. Similarly in a fixed exchange-rate system, the central banks promise to convert their liabilities into foreign exchange at a fixed price. 

The ECB and other modern central banks that are on a floating exchange-rate system make no such promise. As a result, the value of the central bank’s assets has no bearing for its solvency. The only promise made by the central bank in a floating exchange-rate regime is that the money will be convertible into a basket of goods and services at a (more or less) fixed price. In other words the central bank makes a promise of price stability. That’s all. 

Thus it makes no sense to state that the limit to the losses a central bank can make at any point in time is given by the present value of future profits (seigniorage). There is no such limit. The central bank can make any loss provided the loss does not endanger its promise to maintain price stability."

De Grauwe and Ji discuss the situation of a central bank with one sovereign, then discuss a central bank in a monetary union with many sovereigns. Consider the central bank of a stand-alone country buying government bonds in the secondary market:

"Government debt that carries an interest rate and a default risk becomes debt that is a monetary liability of the central bank (money base) that is default-free but subject to inflation-risk. To understand the fiscal implications of this transformation, it is important to consolidate the central bank and the government (after all they are separate branches of the public sector).
After the transformation the government debt held by the central bank cancels out. It is an asset of one branch (the central bank) and a liability of another branch (the government). As a result, it disappears. The central bank may still keep it on its books, but it has no economic value anymore. In fact the central bank may do away with this fiction and eliminate it from its balance sheet and the government could then eliminate it from its debt figures. It has become worthless because it was replaced by a new type of debt, namely money, which carries an inflation risk instead of a default risk. 

This is why it makes no sense to say central banks lose when the market price of the government bonds drops. If there were a loss for the central bank it would be matched by an equal gain of the government (whose market value of the debt has dropped in the same proportion). There is no loss for the public sector... 

When the central bank has acquired government bonds, a decline in the market value of these bonds has no fiscal implications."

What about in a monetary union (that is not a fiscal union) like the Eurozone? De Grauwe and Ji ask us to imagine the ECB buys €1 billion of Spanish bonds with a 4% coupon. Then the ECB would receive €40 million interest annually from the Spanish Treasury and return this €40 million every year to the national national banks according to national equity shares in the ECB. So 11.9% of the €40 million would go back to the Banco de España. The rest would go to the other member central banks, with 27.1% (or €10.84 million) going to the German Bundesbank. In short, the authors say, "An ECB bond-buying programme leads to a yearly transfer from the country whose bonds are bought to the countries whose bond are not bought."

I'm not sure if it is fair to call the example above a transfer away from Spain. Sure, Spain would make an interest payment to the ECB and receive only 11.9% of it back. But without the program, Spain would make an even larger interest payment to other lenders and receive none of it back. The ECB purchase would be a net gain for Spain. And I agree that it would benefit Germany, but not because of the €10.84 million, small change in the scheme of things. The real benefit is compared to the counterfactual of a member state sovereign insolvency and contagious financial instability. Remember that no bonds have been purchased under the OMT so far, so no interest payments have been made, and the hope is that "a credible commitment alone is sufficient to eliminate the speculative equilibrium." Also remember that according to De Grauwe and Ji's earlier logic, if the ECB wanted to give Germany €10.84 million, they don't need the corresponding asset backing of a Spanish interest payment to do so.

So, while I agree with De Grauwe and Ji that German taxpayers benefit from the OMT, and hope for the program to continue, I don't think interest payment transfers are the main point to emphasize. Rather, as Holger Schmieding of Berenberg Bank writes,

"Critics claim that the OMT redistributes risks within the euro zone and that the ECB has no mandate to do so. That is disingenuous. First, the major effect of the OMT is to reduce the level of risk for everybody in the euro zone. An economic depression with a chaotic collapse of the euro would have been much more expensive even for the German taxpayers than any risk that may come with the OMT. Second, due to the OMT announcement, the ECB balance sheet has contracted and improved in asset quality, reducing risks for German taxpayers. Third, it is the very nature of monetary policy to change relative prices and hence risks in financial markets. That is how monetary policy sets incentives for households and companies to adjust their behaviour. Outlawing this feature would mean outlawing monetary policy itself."

In a week from today, I will start working as a graduate student instructor for an undergraduate macroeconomics course at Berkeley. It will be fun to bring up this court case when they are learning about the standard textbook distinction between fiscal and monetary policy and see what they think.

This Time is Not So Different: The Euro Crisis and the 1840s

In the United States, the 1840s were "an era of fiscal crisis following a decade of fiscal exuberance," according to a paper by Arthur Grinath, JohnWallis, and Richard Sylla. This paper was written in 1997, but its insights into a sovereign debt crisis of long ago provide interesting parallels to today.

In the 1820s and 1830s, state governments made large investments in canals, railroads, and banks. New York and Ohio were the first two states to start canal projects. At first, the expected revenue from the projects was low or uncertain, so New York and Ohio raised taxes to service the canal debt. But the Erie and Ohio canals were highly successful, so subsequent canal construction was financed without corresponding tax increases. New York, Ohio, and other states expected internal improvement projects to produce future revenue, so they didn't feel the need to raise taxes when they began new projects. States were easily able to issue bonds to domestic and foreign (especially British) investors to finance their projects:

“Both state borrowers and lenders, foreign and domestic, anticipated that states could tax land if their bank and transportation projects failed. After 1836, increasing land values and taxable acreage were the common factor underlying state fiscal policies, bank investments, and transportation improvements nationwide. Northeastern states knew they had large amounts of untaxed land, rising in value. It was a fiscal reserve against which they could borrow to finance extensions of their transportation systems. Western states, north and south, were in the midst of the greatest land boom in American history. If northwestern states were uncertain about just when transportation investments would generate revenues, they nonetheless anticipated that many more, and more valuable, acres could soon be taxed. States were thus confident that property tax proceeds would provide adequate fiscal resources to service the debts they incurred. Investors in state bonds concurred.”

State government bonds were considered safe assets because it seemed inconceivable that a state government could default-- their investments were expected to be profitable, and even if they weren't, the states had plenty of potential to increase their tax revenue, especially since land value was rising. Grinath et al. quote Illinois Governor Ford as saying "Mere possibilities appeared to be highly probable, and probabilities wore the livery of certainty itself.”

Gary Gorton, Stefan Lewellen, and Andrew Metrick define a safe asset as one that is information-insensitive. "To the extent that debt is information-insensitive, it can be used efficiently as collateral in financial transactions, a role in finance that is analogous to the role of money in commerce." Gorton elaborates on this idea in an interview with the Region magazine, explaining that debt is "easiest to trade if you’re sure that neither party knows anything about the payoff on the debt." In other words, "The depositors believe that the collateral has the feature that nobody has any private information about it. We can all just believe that it’s all AAA."

In the 1830s, both the states and the investors in state bonds could "believe that it's all AAA" since, even if investment projects turned out not to generate much revenue, states had seemingly boundless untapped tax potential. Thus it was unnecessary for investors in state bonds to find information about the details of states' particular projects. State debt was information-insensitive, a useful property considering how slowly information traveled across the Atlantic in those days.  No need to calculate probabilities when probabilities wear the "livery of certainty itself."

Gorton says that the few really big crisis events in history come from a regime switch in which debt that is information-insensitive becomes information-sensitive. This is precisely what happened in the U.S. states. During the early 1830s expansion and boom of 1835, state debt was information-insensitive, especially as ever-rising land prices promised a large and growing fiscal reserve. But as several domestic and external factors combined to bring about the panic of 1837 and collapse of 1839, the strength of the fiscal reserve was challenged. As land values and property taxes fell, the quality of state's canal, bank, and railroad investment projects suddenly mattered for their ability to service their debt. The situation is described in another paper by Wallis and Namsuk Kim:

In July of 1839, the Morris Canal and Banking Company of New Jersey defaulted on Indiana, and the state quickly was forced to curtail construction on its network of canals and railroads. By the autumn, Illinois and Michigan were forced to slow or stop construction when investment banks defaulted on their obligations to the states. Land sales and land values in these northwestern states had been rising steadily through the 1830s. When transportation construction stopped, land values and property tax revenues began falling and, by late 1839, it was apparent that these states would soon have trouble servicing their debts. In January
of 1841, Indiana was the first state to default on interest payments.

It was a nasty spiral-- as infrastructure projects failed, land values and tax revenue fell further, eroding the states' fiscal positions, making it harder for them to issue bonds and forcing them to pay higher interest rates. This further deteriorated their fiscal positions, and led to suspensions of infrastructure projects and yet higher interest rates. This is similar to what happened in the eurozone, for example in Greece. In the early 2000s, Greece was able to run large deficits without facing high borrowing costs, because the growing economy made Greek sovereign debt information-insensitive. The economic crisis was a "regime change" making sovereign debt information-sensitive. Without the benefit of a fast-growing economy, the Greek government's ability to pay depending much more on its fiscal position, so borrowing rates rose, causing an even worse fiscal position.

The parallels with Greece continue. Many states found, when they tried to raise taxes, that they lacked the state capacity to do so. Property taxes were extremely politically unpopular, and states had trouble not only passing tax legislation but also implementing tax collection. In Maryland, for example, three counties refused to remit their share of the property tax imposed in 1841, and seven refused in 1842. It wasn't until 1845 that the tax was effectively implemented, allowing Maryland to resume debt service. Other states were even less successful in raising property taxes and ended up defaulting and repudiating their debt. Greece also faces a tax evasion problem and encountered serious public and political opposition to attempted austerity measures. In an earlier post I mentioned a paper by Mark Dincecco and Gabriel Katz called "State Capacity and Long Run Performance." State capacity refers a state's ability to tax and to provide public goods and services, called its extractive and productive capabilities, respectively. Dincecco and Katz write:

We argue that the implementation of uniform tax systems at the national level – which we call “fiscal centralization”– enabled European states to effectively fulfill their extractive role. This transformation typically occurred swiftly and permanently from 1789 onward. Similarly, we argue that the establishment of parliaments that could monitor public expenditures at regular intervals – called “limited government” – enabled them to effectively fulfill their productive role. This transformation typically occurred decades after fiscal centralization over the nineteenth century. By the mid-1800s, most European states had achieved “modern” extractive and productive capabilities, implying that they could gather large tax revenues and effectively channel funds toward non-military public services. We argue that these critical improvements in state capacity had strongly positive performance impacts. 

The institutional changes that Dincecco and Katz describe in the late 18th century Europe that brought about extractive capabilities include fiscal centralization and parliamentary, limited government. The U.S. states in the 1840s, and apparently some of the European states today, lack such state capacity, a fact which plays a role in the crises then and now.

Pennsylvania's canals were a financial disaster, so the state faced particularly high borrowing costs, until the Bank of the United States was rechartered as the Bank of the United States Pennsylvania (BUSP). The bank's charter included a promise to underwrite $6 to 8 million in state bond issues. The bank agreed to lend to Pennsylvania at 4%, and as a result, Pennsylvania bond yields in Philadelphia stayed very near to 4% for the next few years. Wallis and Kim write, "Deliberately or not, the BUSP pegged the price of Pennsylvania bonds as a result of its obligations to purchase state bonds over this 18-month period." This is similar to the ECB's Outright Monetary Transactions (OMT) policy, which, by promising to buy sovereign bonds of Eurozone member states, aims to bring down bond yields and lower borrowing costs for countries that face problems selling debt. Though the bank loan program in Pennsylvania was temporarily successful in helping Pennsylvania borrow at lower cost, when the BUSP closed down in 1841, Pennsylvania bond yields jumped immediately, from 6.01% in January 1841 to 9.5% in March.

What ultimately happened in the United States was that the debt crisis forced a change in the structure of public finance. States initiated constitutional restrictions on debt issue and instituted requirements that new spending be matched by new tax increases. The debt crisis in the euro area is also likely to change the structure of public finance, but not in the same way. The United States is both a monetary union and a fiscal union, so even though the states adopted balanced budget amendments, the federal government could still do countercyclical fiscal policy. The euro area is a monetary union without a fiscal union, so it would be very costly for states to institute such restrictions on deficit spending. One possibility is that the euro area will become more of a fiscal and/or banking union; or there may be other changes in the structure of public finance that I can't foresee.

Europeans' Biggest Problem

The European Commission's Eurobarometer survey monitors public opinion on a variety of political and economic issues across European Union member states. One question on the Eurobarometer survey asks:

Personally, what are the two most important issues you are facing at the moment? 

This question was only asked in May 2012. For the EU as a whole, by far the most common response was rising prices/inflation. In fact, 45% of people in 2012 said that inflation was one of the top two most important issues they were facing. The pie graph below shows, for the EU as a whole, the responses people chose. Only 15% of people chose the financial situation of their household as a top issue. Health and social security also had a mere 15%. I was stunned that three times as many people consider inflation a top issue as consider health and social security a top issue.

In the graphs below, the results are broken down by country. First I show the percent of respondents in each country who choose inflation as a top-two issue. Then for a few countries, I show the percent who choose inflation and the percent who choose unemployment. In twelve countries (including Austria, France, and Germany), at least half of respondents say that inflation is a top-two issue. Sweden is a major outlier-- only 5% think that inflation is a top-two issue. The next lowest is Greece, at 26%. Sweden and Greece did have the lowest inflation in the EU in May 2012, but really just about ALL countries in the EU had (and still have) low or reasonable inflation.

Half of Germans and French thought that rising prices were a top issue, even when inflation was just 2.5%. The EC Consumer Survey, asks people how much they think prices have risen in the past 12 months. In May 2012, 28% of German, 36% of French, and 40% of Austrians thought that prices had risen "a lot."

Neil Irwin recently wrote that "The leading economies of the industrialized nations may not have a lot in common, but they are all afflicted by this: Inflation is too low." Even though inflation is too low, a lot of people think it is high-- and think that rising prices personally affect them more than unemployment. Public opinion is a powerful force, so we see policymakers being more reluctant to raise inflation when it it too low, than to lower it when it is too high.

Trust and the Future of the Euro

Last week I attended the Future of the Euro conference at UC Berkeley. The conference was cosponsored by the Institute of European Studies, the EU Center of Excellence, the Austrian Marshall Plan Foundation, and the Austrian National Bank. The conference was structured around four panels on political, banking, fiscal, and monetary union. On each panel, several panelists--mostly economic historians-- gave brief presentations of about 20 minutes. In each case, I wished I could hear longer presentations to get more details. This was particularly true of the presentation by Lars Jonung from the monetary union panel, in part because he gave a critique of modern macro which I hadn't heard before.

Jonung is a senior professor in economics at Lund University, Chairman of the Swedish Fiscal Policy Council, and Research Adviser at DG ECFIN at the European Commission. Jonung says that when applying macroeconomic models, we need to keep in mind economic political culture, which he claims can be summarized as “trust.” The indirect cause of the euro crisis, he argues, was lack of trust by the public in the political system in certain parts of the Eurozone, especially Greece, Spain, and Cyprus. He describes the “vicious trust circle” in those areas, whereby weak performance leads to low trust, which leads to more weak performance, and so on. In other parts of the Eurozone, such as Germany and the Nordic countries, a “virtuous trust circle” is in place.

Jonung uses the Eurobarometer dataset as an indicator of trust. He says that the Eurobarometer ought to be a gold mine for researchers, but is rarely used because modern macro models include no role for trust. From his own analysis of the Eurobarometer, he finds that trust in national governments, the European Parliament, and the European Central Bank have fallen over the 1999-2012 period, but support for the euro has not. Jonung remarked, “People trust the euro, but they don’t trust the institution behind the euro.”

His proposed long-run cure for the euro area is to foster trust in national governments and the EU system via decentralized fiscal policy. For starters, this would entail fiscal policy councils in every member country, as well as Eurostat offices in every member country. (See a paper by Jenny E. Ligthart and Peter van Oudheusden on the role of fiscal decentralization on trust in government.) If the recovery continues to be slow, so that trust cannot be restored across the Eurozone, he thinks that one possibility is to have a “high trust monetary union” in the North and a “low trust monetary union” in the South, with flexible exchange rate between the two. Then, when trust is equalized, they could fix the exchange rate.

As I mentioned, I wished Jonung’s talk could have been longer, because it left a lot of missing details. First, I would have liked to hear more about his high trust and low trust monetary unions idea, particularly, how trust would eventually be equalized between the two unions. Much more fundamentally, I’m not sure exactly how Jonung interprets the concept of trust. The Eurobarometer survey asks respondents a large number of questions about their trust in, and support for, the euro and various institutions. But what do respondents mean when they say that they trust a currency or institution? Does trust in a government mean confidence that the government is not corrupt? Does it mean approval of government’s policies toward people in their situation, or in all situations? Does it mean optimism that the government will be successful in preventing hyperinflation or recession? Does it imply that the government is credible, or something else? What about trust in the euro? Does it simply measure the subjective probability that the euro will persist? Is it a sense of European identity? On the one hand, I am sympathetic to the argument that macroeconomic models do not incorporate political cultural factors like trust. On the other hand, depending on what exactly is meant by trust, it may be reflected in interest rates, exchange rates, and other features of the asset market which are to some degree built into macro models.

Jonung's recent working paper with Felix Roth and Felicitas Nowak-Lehmann D. notes that when respondents are asked about "trust" in the euro, "It seems reasonable to interpret 'trust' in the euro as `trust' in the purchasing power of this type of money," but this is not what the authors mean by trust. Instead they focus on the question about "support" for the euro, which "would then mean support for the idea of a single European currency while not necessarily meaning that the respondent expects the euro to deliver a stable purchasing power." This is more what the authors mean by trust. 

The first figure below comes from the Roth, Jonung, and Nowak-Lehmann's paper showing support for the euro in the EA-12 countries (where support is their indicator for trust.) I made the second figure at the Eurobarometer website (European Commission Public Opinion). It shows, for Greece only, the percent of respondents that tend to trust or not trust in the European Union. You can see trust in the EU fall drastically around crisis time. But I'm not sure exactly what this tells us, since respondents could interpret the question about trust in a variety of ways that would all show a decline in trust following a crisis. The fact that trust continues to fall, and hasn't leveled off, does seem interesting.

You can play around on the Eurobarometer site and see how trust in the EU and ECB changes in different countries with different country-specific or Europe-wide events. Roth, Jonung, and Nowak-Lehmann "interpret the fall in trust in the ECB to imply that citizens blame the ECB for not preventing the economic, financial and political turmoil during the crisis and suspect that the crisis measures taken by the ECB and other European institutions have had an inflationary effect," while at the same time, "respondents support the euro as their currency and...do not blame the euro for the crisis."