The Network of Risk

MET Administrative Sciences faculty take on the financial crisis and bring it back to class.

The FDIC Failed Bank Lists show that about 25 commercial banks failed between 2000 and 2007. That number rose to 350 in the period from 2008 to 2011. What happened?

“We allowed banks to securitize mortgages and start trading them like stocks,” says Associate Professor and Associate Chair of Administrative Sciences John Sullivan. In 2008, the U.S. housing bubble burst, leading to an increase of subprime mortgage delinquencies and foreclosures—and a sharp decline in the value of securities held by global financial institutions deemed “too big to fail.” The shock was felt throughout the world economy. As Sullivan explains, “We’re all interconnected now. When the United States gets sick, Europe catches a cold.”

The high degree of consolidation that has occurred in the banking system creates significant issues, agrees William Chambers, an associate professor of the practice of administrative sciences who worked at Standard & Poor’s for twenty years. “The top ten banks control a majority of banking assets and deposits. It makes those ten banks essentially too big to fail (or as attorney General Eric Holder has said, ‘too big to jail’).”

The study of interdependent financial and economic networks is the forte of Assistant Professor of Administrative Sciences Irena Vodenska (UNI’09)—a Chartered Financial Analyst (CFA) and former investment banker whose background includes portfolio management, financial analysis, and securities trading on Wall Street and European markets. Vodenska explains that the global financial crisis spurred a cascade of events throughout the interconnected global economy, propagating value deterioration of most financial markets around the world; contributing to the potential collapse of major financial institutions; miring governments in the bailout of too-big-to-fail banks; adversely affecting the global housing market; contributing to increased unemployment rates and prolonged unemployment; significantly reducing consumer wealth and appetite for spending; and essentially precipitating the European sovereign debt crisis.

“We’re all interconnected now. When the United States gets sick, Europe catches a cold.”

Vodenska is the principle investigator for Boston University’s part of the international, multi-institutional “Forecasting Financial Crisis” grant funded by the European Union. Her research applies complex network theory to predict systemic risk in financial systems, and is conducted in collaboration with BU Professor of Physics and William Fairfield Warren Distinguished Professor Gene Stanley, doctoral students from BU’s Department of Physics, and master’s students from MET. Published this past February in Nature’s Scientific Reports, “Cascading Failures in Bi-partite Graphs: Model for Systemic Risk Propagation” describes a model that could be useful for systemic risk stress testing in banking networks and interconnected economies such as the EU—especially critical in the aftermath of the European sovereign debt crisis.

“We focus on quantitative methods for modeling volatility, global interdependence of financial markets, banking system dynamics, and studies of extreme events such as bubbles and crashes,” says Vodenska. “With the recent financial crisis, we saw the need to analyze the interplay between the financial and real estate sectors of the economy, and to consider how real estate investments are financed, the consequences of economic boom-and-bust cycles, and the nature of the bank assets.

“We want to develop a good understanding of the dynamics of these networks and ideally create an early warning system,” continues Vodenska. “If we can detect risk increases in the network, we can alert policymakers. The idea is to make our models applicable to policymaking, so they can be used by policymakers to intervene before a major crisis happens.”

“If we can detect risk increases in the network, we can alert policymakers.”

Vodenska indicates that the crisis has enhanced her courses by motivating the “expansion of fundamental theories” and opening new frontiers in research.

It is opportune, then, that this will be the first year that all full-time faculty have access to research assistants. “They will support faculty while being exposed to real financial situations. It brings the learning from research back into the classroom,” says Chambers. “In the wake of the financial crisis, many basic assumptions about the economy and the financial system have been challenged. It’s a very interesting time to get involved in the field and analyze what’s going on.”

Sullivan, who has many years of experience in mergers and acquisitions, spends the first half hour of class discussing the realities of the market: “One of our goals is to teach practical finance. Don’t get caught up in bubbles, look at the fundamentals, and think analytically. But it’s not a science—it’s an art. Nobody knows what is going to happen tomorrow. We make educated estimates, and consider how to minimize risk and maximize returns.

“If you want to learn economics, this is your program,” concludes Sullivan. “You’re going to get the best of both worlds: an applicable degree taught by former executives who have been in the water and know how cold it can be.”

Select a MET professor to find out more about his or her views on the financial crisis and learn about the financial programs offered at MET.

  1. Associate Professor of the Practice William Chambers

    William Chambers

    Associate Professor of the Practice

  2. Associate Professor and Associate Chair John Sullivan

    John Sullivan

    Associate Professor and Associate Chair

  3. Assistant Professor of Administrative Sciences Irena Vodenska

    Irena Vodenska

    Assistant Professor
     

William Chambers

Associate Professor of the Practice of Administrative Sciences

PhD, MPhil, MA, Columbia University; BA, College of Wooster

Dr. Chambers has over twenty years of experience developing and overseeing credit models, internal credit scoring systems, and default risk assessment processes for Standard & Poor’s. He is an expert in international finance, portfolio management, and the economics of real estate development. Chambers teaches undergraduate and graduate courses in finance, investment analysis, portfolio management, and financial markets and institutions.

Metropolitan: What are the main strengths of MET’s programs in banking and finance? How do we prepare students for tomorrow’s economy?

Chambers: We want our graduates to be able to go to work for companies, fit in, and be productive very early on. We want to give them the kind of tools, knowledge, and framework for understanding problems so that they can jump into any situation presented to them. Therefore, we approach things from an applied standpoint. All of our courses involve application, not just pure theory—looking at how events actually take place on the ground in the investment, financial, and business worlds. All of our instructors have a significant amount of real-world experience. I spent the vast majority of my career working in the securities and consulting industries. John Sullivan brings many years of experience in mergers and acquisitions—not just the theoretical framework of that, but their actual implementation. Irena Vodenska brings a lot of experience in terms of portfolio management. That real application adds a new dimension to the program.

A second element is that every part of business has a financial aspect to it, whether it’s a capital budget or pricing decision, a merger, acquisition, or a disposal. Therefore, students need to learn how financial factors interact with other parts of the business. Our students don’t take twelve finance courses; they have a core of courses that illustrate—along with the specialized finance courses—how the tools they learn in finance can be applied in various areas, from product development and innovation to e-commerce marketing. The way we’ve designed the program and the way we’ve tried to present it makes it more valuable to the students and their future careers.

In a nutshell, how would you characterize the state of the global economy today?

The economy is growing in a very gradual way: two steps forward, one step back. There is no single area of the world that’s driving growth at this time. China, which had been growing at rates upward of ten percent per year, has slowed to six or seven percent per year. Other Southeast Asian countries are growing, but in a much less moderate fashion. Europe is the weakest part of the globe at this point, in terms of economic growth. Latin America is showing some signs, but, again, it’s two steps forward, one step back. The U.S. economy is recovering from the recession, but very slowly and much more slowly than after almost any other recession we’ve encountered, partially because the usual driver for recovery in the U.S.—the housing market—is one of the main reasons for going into the recession. The lack of stimulus in the housing sector has had reverberations in much of the rest of the economy. It’s recovering and growing—just not as much as we would like it to.

Is this financial crisis different from others in the past?

The joke in finance is you can answer any question with “it depends.” This crisis is unique in some ways, yet also very much the same as others we’ve seen. Any student of economic history can trace the whole trend of financial bubbles: the uptrends, the bursting, and then the recovery. We’ve seen those patterns over and over again, whether we talk about the tulip mania in the Netherlands several hundred years ago, the South Sea bubble here in the U.S., the buildup to the Great Depression in the 1920s, or a range of other things that have occurred in various periods of time. If we go back into the ‘80s, we see huge investments in Latin America and then disappointment and significant losses there. So we’ve seen significant bubbles and the bursting of those bubbles over time; I just think that sometimes we have very short memories.

As for the unique aspects: The degree to which the expansion took place, and how long it was able to continue, certainly made the fall worse than it would have otherwise been—and caused greater reverberations. I think almost everyone in 2006 or 2007 would have said there was a housing bubble, though they might have argued about how much and how widespread it was. Instead of leveling off, we saw huge reductions in housing prices, which in some parts of the country exceeded even those of the Great Depression on a percentage basis. So, this recession was special in some ways. The fact that it was global is another real distinguishing factor from other recessions and bubbles that affect just one country or one region.

What kind of lessons do you think we learned?

One of the interesting things we’ve seen in finance in general is the challenge of the argument that all decisions are made on a purely rational basis. That’s a fundamental assumption in both economics and finance: intelligent people will make rational decisions. But we’ve seen a whole movement towards behavioral economics and finance that identifies variances from that assumption. For example, people tend to look for information that confirms their preconceived notions and ignore information that contradicts those notions. That doesn’t sound like a great revelation, but that has a great effect on what we identify as a “reasonable” answer. We often assume that the status quo will continue into the future and that any change from that has a very low probability. The whole aspect of decisions being made on a rational basis has been challenged very dramatically and illustrates some of the results we’ve seen in the financial crisis.

A second thing is a challenge to the belief that totally free markets can be self-regulating. We’ve seen that that doesn’t work nearly as well as many thought it would. People and institutions will take risks well beyond what they should be taking and focus on possible short-term gains, ignoring medium- and longer-term risks. Certainly some aspects of this have long been identified, such as the so-called “agent problem”—namely, people working for a company who are interested in maximizing their own well-being above the company’s well-being. If they can take additional risks and get a bigger bonus, then they’ll take additional risks, perhaps well beyond any that the company, which will have to foot the bill at the end, is comfortable with. So, compensation schemes have to be re-examined.

We take these theories and conclusions and integrate them into the courses we teach. Being aware of the issues is very valuable.

John Sullivan

Associate Professor and Associate Chair of Administrative Sciences

PhD, Northeastern University; AM, Harvard University; MBA, Northeastern University; BA, Regis University

Dr. Sullivan is an expert in health care policy and finance and frequently appears as a commentator in local and national media. He teaches mergers and acquisitions, corporate finance, capital markets, investments, and new business ventures. He served as senior analyst for corporate development at Fresenius Medical Care, and has provided strategic consulting for various health care organizations.

Metropolitan: From an educational perspective, is it a particularly interesting time to study banking and finance?

Sullivan: It’s a very interesting time to study finance because of everything that’s going on with the global economy, as well as the “super banks” trying to recover. Financial stocks are way up because of the belief that they’re actually turning the corner. Interest rates are about zero. The treasury is propping up the U.S. economy. Europe is a fascinating place—we can’t get our own government to work together, let alone seventeen countries. We’re healing, but it’s going to take time. The danger is going to be inflation—when does it kick it? You can’t have interest rates at zero for a prolonged period of time without either causing stagnation or inflation. The stock market is up, and my theory on that it’s up only because there’s nowhere else to put your money. You can’t put it in the bank. You can’t buy debt, because that’s too expensive. Your yield is close to zero. I don’t think the stock market is a bubble, but if you see inflation, maybe it’s time to move your money out of the market.

Is this financial crisis different from others in the past, and if so, why?

It’s different than the Great Depression, in which the government didn’t intervene. Despite some of the critics, the government stepped in this time, which I think was the right thing to do. If they hadn’t, the crisis would have been far worse. There would have been runs on banks, and basically you would have duplicated the Great Depression. I think Dr. Bernanke did the right thing in stepping in, propping up the banks, and restoring confidence. Even though we took a hit on the chin, I think it would have been far worse had the government not intervened. If the banking system was allowed to collapse like Lehman, it would have been far, far worse of a recession, almost depression-like. It would have been worse than what we saw in 1929.

What kind of lessons do you think we learned?

Well, I think—and this is still going on—we allowed banks to securitize mortgages. When you securitize mortgages and start trading them like they’re stocks, then that is a problem because you don’t care if they default or not. In the old days, what would happen is that if you wanted to buy a house, the lender would scrutinize your income and your ability to pay, they would loan you money, and would hold that mortgage and then give you the deed when you paid it off. If the bank can trade this mortgage, they don’t care if it defaults or not. I think that’s a lesson that we have not learned.

My father had a saying: “You don’t put locks on doors to keep criminals out. You put locks on doors to keep honest people honest.” What you do is remove the temptation to push the envelope. We’ll see. I don’t know if we’ve really learned our lessons about the meltdown. It would be easy to fix: stop securitizing loans. You reinstall the Glass-Steagall Act, which was dismantled and allowed the banks to merge to create “too big to fail.”

In what ways do current events in the global economy—such as the financial crisis—inform the content of your courses?

At the beginning of every class, I spend a half an hour discussing what’s going on the market and why it’s happening. Some of it is counterintuitive, some of it is predictable. That’s the great thing about finance, that every day is different. I think we’re lucky in that the faculty here can anticipate things. We’ve risked our own careers, have analyzed markets, and have done transactions. One of the misconceptions about finance is that it’s a science, just because there are numbers involved. It’s not a science, it’s an art. Anyone who would disagree with me, I would ask “What’s the market going to do tomorrow?” Nobody knows. If you do know, please tell me. While we do have some scientific tools to look at what’s going on in industry, we don’t know tomorrow’s going to bring. We make educated estimates as to what’s going to happen tomorrow and I think that’s part of what makes finance so fun, as opposed looking back and counting. I don’t care about yesterday; what’s going to happen tomorrow? Where’s my risk, and how do I minimize it and maximize my returns?

Irena Vodenska (UNI’09)

Assistant Professor of Administrative Sciences

PhD, Boston University; MBA, Vanderbilt University; MS, BS, University of Belgrade

Dr. Vodenska is an expert in international finance and investments, with more than fifteen years of hands-on experience in financial analysis and securities trading on Wall Street and European markets. She is a Chartered Financial Analyst with experience in creating and actively managing hedge funds, specializing in risk arbitrage and convertible fixed-income securities. Vodenska has broad experience in academic teaching and corporate training, and her research interests include statistical finance, applications of quantitative methods in financial modeling, and interdependent financial and economic networks, an area closely related to econophysics, network theory, and complex systems.

Vodenska was a member of BU’s PhD thesis committee and the second reader of “Network Theory and its Applications in Economic Systems,” the dissertation of Graduate School of Arts & Sciences physics student Xuqing Huang (GRS’13), who graduated this past May.

See an interview with Vodenska by the FOC project (Forecasting Financial Crisis).

Metropolitan: Describe your recent paper published in Nature’s Scientific Reports.

Vodenska: With the most recent financial crisis, my research collaborators and I saw the need to analyze the interplay between the financial and real estate sectors of the economy, and to consider how real estate investments are financed, the consequences of the economy boom and bust cycles, and the nature of the assets of the banks. We model interconnected financial and economic systems as complex networks to study the interactions between them and investigate systemic risk propagation from one financial system to another.

Our paper, “Cascading Failures in Bi-partite Graphs: Model for Systemic Risk Propagation,” addressed the problems in the U.S. financial network. The interesting outcome was that we didn’t find evidence that the residential mortgages were the main culprits of the real estate bubble burst in the United States—it was really the commercial real estate mortgages. We found that our model could identify a significant number of the actual failed banks in the United States. The nice part about this research is that we had the actual data—the U.S. bank balance sheet data, which is publicly available. We also had the Federal Deposit Insurance Corporation (the FDIC) Failed Banks List data.

From 2000 to 2007, we saw 27 or so banks fail. From 2008 continuing through 2012, over 400 banks failed, so the test data was extensive and we could test our model on it. We could identify failed banks with pretty high accuracy at the true-positive rate of 78 percent, at an expense of close to 20 percent false-positive rate (meaning they were identified as failed, but they did not fail). What we learned is that the commercial real estate assets were the major culprits responsible for the real estate bubble burst.

You are applying your findings as principal investigator for BU’s part of the international, multi-institutional “FOC Project (Forecasting Financial Crisis)” grant, funded by the European Union. Is the scale of this project unusual because of its international nature?

I think it’s very unique anytime you’re able to put so many institutions together in a research environment. Funding from the EU is really a very big catalyst to enhancing interdisciplinary and international collaboration, because it allows for the exchange of ideas on common projects. Collaboration is required because we have to report to the EU—we have to address certain problems.

It’s very difficult to establish these collaborative relationships in research, so we now value them so much that even when this project is over, we will continue to work on other projects.

How would you characterize the state of the global economy today?

Uncertain and too connected.

Look at the Greek and Irish debt crisis. The Greek GDP is only $300 billion; the Irish GDP is only $200 billion. Combined, that’s smaller than the GDP of Pennsylvania. So, why is the impact so great if these economies are so small? One plausible answer is interconnectivity, or the existence of a “network of networks.”

Major financial institutions have invested in the sovereign debt of Greece, Ireland, Spain, Portugal, or Italy. When the Greek debt became more risky, its value reduced and banks wanted to sell it. Once they started selling, the price deteriorated, and then other banks wanted to sell—which exacerbates the effect. Reduction in value of the Greek debt precipitated a reduction of the value of the bank as an entity in the banking network. So, then this bank will sell its Italian debt. Now the Italian debt is going to reduce in value, and this goes into a feedback loop with the other banks, and a cascading failure happens.

Italy and Spain are much larger economies. Italy has close to €2 trillion debt outstanding with 50 percent financed externally. Spain has over €700 billion of public debt outstanding and an unemployment rate of over 22 percent. This coupling of economic problems could lead to a much larger cascading failure of economic systems. It’s much more dangerous compared to the dynamics of a single network.

This is the reason my collaborators and I are engaged in researching the network of networks, or coupled networks, or multiplex networks.

How might your research impact the issue of too much interconnectivity?

Our research helps us to improve our understanding of the dynamics of these financial networks, enabling us, ideally, to create an “early warning system.” We developed a mathematical framework to study the stability of interdependent networks under distress and look for predisposition of selected nodes to dominate in distress propagation through the network. The main objective of our contribution to the FOC project is to propose safety measures that can be useful in preventing cascading failures in economic networks—or, at least, soften risk propagation effects on different economic networks of networks. Specifically, we focus on modeling dramatic damage to the system, including phase transitions—which is a statistical physics concept—in order to understand how to prevent coupled networks from switching from seemingly stable to irreversibly unstable states.

In what ways do your students get involved in research at MET?

Mostly, students get involved as research assistants. They work on research projects in quantitative analysis and statistical modeling, collecting and analyzing the data, and doing the preliminary analysis. The feedback from students is that this helps them when they’re looking for jobs, because—by discussing empirical research projects—they can demonstrate to potential employers that they can think creatively about solutions to real problems in finance and economics.