Leendert Pieter de Neufville was only 21 in 1751 when he founded his Amsterdam bank, De Neufville Brothers. His big chance came a few years later with the Seven Years’ War — what Americans call the French and Indian War.

Amsterdam was then the most sophisticated financial center in the world, leading in financial innovation and having the most creative bankers and the highest amount of speculative capital. The Dutch guilder was the U.S. dollar of its day, the reserve currency that facilitated trade throughout Europe and beyond.

De Neufville took full advantage, building up one of the world’s wealthiest and most prestigious banks by financing the Prussian side of the war. He lived well, furnishing his house only with the finest-quality objects and owning an excellent collection of paintings — but not a single book.

The way he made his money was thoroughly modern: rapid, irresponsible financial innovation in the form of acceptance loans, not all that different from the financial instruments so damaging in the 2008 crisis. Cheap short-term borrowing was used to make long-term loans at high interest rates involving long chains of obligations, spanning multiple banks and countries.

The key to all the profit was borrowed money. For every 23 guilders he lent to Prussia, De Neufville supplied one and borrowed 22 secured by commodities. Highly profitable in good times, but it didn’t take much for things to go wrong, as they did spectacularly when the war ended in 1763, culminating in the first modern financial crisis.

Commodity prices crashed because the farmers could now finally start producing again, making all the commodity-based collateral behind De Neufville’s acceptance loans worth little. Investors got spooked and decided not to roll over these short-term loans — they all went on strike, just like their successors in 2007. As De Neufville didn’t have enough ready cash to repay his creditors and keep his bank alive, he had to sell his vast holdings of commodities. However, that only caused prices to fall further, a process known as a fire sale. Falling prices induce speculators to sell, and when they sell prices fall further in a vicious loop. It did not take long for De Neufville to default.

Thus was born the first global systemic crisis, described at the time as a pest epidemic. It spread with raging speed from house to house. The city of Hamburg was hit hard, and on August 4 its mayor wrote to the mayor of Amsterdam asking him to bail out De Neufville. Amsterdam refused.

Fortunately, the crisis turned out to be relatively short-lived, and Amsterdam, Hamburg and the other major centers recovered quickly. Berlin suffered badly because of how its emperor, Friedrich II, reacted. He imposed a payment standstill and bailouts, violating the contracts that allowed funds to flow from Amsterdam to Berlin and causing the bankers to mistrust the Prussian authorities. The money stopped moving, and a severe recession ensued. Not the only time a government’s response to a crisis made things worse.

The ABCs of a Crisis

The 1763 crisis is what might now be classified as a V-shaped crisis, except for Berlin, which suffered a U crisis. In a V crisis we see a sharp temporary drop in economic activity, but as nothing fundamental is destroyed the recovery is similarly quick. The Asian crisis in 1998 was of the V variety, while the Great Depression and the 2008 crises, given their sharp crash and slow recovery, were U shaped. What about Covid-19? The jury is still out, but it seems closer to K — quick winners and losers.

De Neufville’s creditors tried to claw their money back but with limited success. He was forced to auction off some of his paintings in 1765, including The Milkmaid by Johannes Vermeer for 560 guilders, now one of the finest attractions of the Rijksmuseum in Amsterdam. Leendert managed to retain most of his houses, including one at Herengracht 70–72 in Amsterdam, a very desirable address then as now.

What happened in 1763 was a systemic financial crisis, and the chance of that happening is systemic financial risk. For the remainder of the book, I will omit the word “financial” and just call it systemic risk. This was the first modern global financial crisis because it was not a crisis caused by war or crop failure, but rather by shadow banking [unregulated informal banking] and the extensive use of sophisticated financial instruments allowing risk to hide and spread by efficient and interconnected financial centers.

We have experienced many systemic crises since, and it is remarkable how similar they all are. Even though the world in 1763 was very different from that in 2008, the crisis dynamics were eerily similar: shadow banking creating deep vulnerabilities that became visible only when it was too late. Crises are all fundamentally the same. They differ only in details.

My definition of systemic risk is the chance that the financial system will fail spectacularly in doing what it is supposed to. The global authorities, in the form of the International Monetary Fund, the Bank for International Settlements and the Financial Stability Board, prefer a narrower definition: “The disruption to the flow of financial services that is (1) caused by an impairment of all or parts of the financial system, and (2) has the potential to have serious negative consequences for the real economy.”

The key here is “financial system.” It has to be either the root cause or the amplifier for a crisis to be called systemic. Since the financial system is so fundamental to the economy, most economic crises are also systemic crises. What about Covid-19? It certainly hit the economy, shaving 10-20 percent off most countries’ GDP in the second quarter of 2020. But the financial system was a bystander, not the cause, and it made things neither better nor worse. The Covid-19 crisis was not systemic.

The focus of systemic risk is not on any individual financial institution; instead, it is on the financial system in its entirety and how it affects the real economy. The failure of a bank, or even a banking crisis, is not necessarily systemic. We need a connection between the financial system and the real economy.

GDP or more, so for the United States in the trillions of dollars. Fortunately, they are not frequent, and I think most people can expect to suffer such a crisis at most once in their lifetime. If one takes the relatively loose definition used in the IMF crises database, then maintained by Luc Laeven and Fabián Valencia, we find that the typical OECD country suffers a systemic crisis once in 43 years on average.

While one in 43 years is the historical average, it is hotly debated whether the future will be as tranquil. A lot of commentators maintain that increased complexity and interconnectedness make financial crises more frequent. The United States and (especially) the United Kingdom are more crisis-prone, the UK enduring a systemic crisis once every 17 years. The last one was in 2008, so 2025 is the due date for the next one if it arrives on schedule.

If anything, the once-in-43-years figure is an overestimate, as the database includes relatively non-extreme events like Black Monday in October 1987 and the Long Term Capital Management crisis in 1998. If we exclude all the mildest crises, we get about one systemic crisis per typical lifetime.

The reason for the once-in-a-lifetime frequency is that the most severe crises happen only after we forget the last one. Crises change behavior, and those who come of age during one will be affected by it for the rest of their lives. We had to wait for the 20-year-olds of 1929 to retire in the 1970s for the seeds of the next crisis to be sown. It then took a quartercentury for the seeds to bear fruit, culminating in the events of autumn 2008. Politics and lobbying push toward a high-risk, high-return financial system, and when memories of the previous crisis fade there is little pushback.

Considering the once-in-a-lifetime frequency of systemic crises, the term “systemic crisis” is as overused today as it was underused before 2008. Most of this usage is imprecise and contradictory, and one often gets the impression that commentators are talking only about the last crisis or financial scandal when they use the phrase.

The financial system is vulnerable to many types of shocks, some coming from outside the financial system, like Covid-19, and others generated by the system itself. Some shocks are idiosyncratic, affecting only a single institution or asset, while others impact the financial system as a whole plus the real economy. The small shocks often arrive from outside the system; all the large ones are created by the interaction of the human beings who make up the financial system. There may be an outside trigger, but the real damage is caused by the system turning on itself.

Anatomy of a Crisis

If I had to describe a textbook financial crisis, it would go something like this: financial institutions have too much money to lend. When they run out of high-quality borrowers, they start making increasingly low-quality loans, often in real estate. In the beginning it all looks ingenious. Developers borrow money to build new houses, which stimulates prosperity and demand for homes. Property prices go up. Everybody feels wealthier and more optimistic, thereby encouraging more lending and more building in a happy, virtuous cycle. This caused many a crisis, like the savings-and-loan debacle in the United States in the 1980s and the Spanish crisis in 2010.

Eventually, when the little boy yells “The Emperor has no clothes!” people realize the prosperity was built on sand. There is no strong underlying economy, and it all comes crashing to the ground; the virtuous feedback loop becomes vicious. Prices fall, developers fail, the banks lose money, the economy contracts, prices fall more — the same fire sale process De Neufville endured. The pre-crisis rise in prices is much slower than the fall: prices go up the escalator and down the elevator.

Most crises follow the textbook. The one in 1914 was atypical, which is why it is my favorite. It was triggered by the assassination of Archduke Franz Ferdinand of Austria on June 28, 1914, leading to posturing among the great powers of Europe amid raised expectations of an impending war. This anticipation created worries that financial institutions would experience difficulties in having crossborder loans repaid — after all, if two countries are at war, enforcing contracts across borders would naturally be difficult. Surprisingly, even after the war started, remittance between the Central Powers and the Allies continued via neutral countries, notably Switzerland, which helped propel it into the ranks of major financial centers.

The immediate consequence of the assassination of the archduke was difficulty in clearing trades. When one buys and sells stocks, it takes time for ownership to be transferred to the buyer and money to the seller, a process called clearing. In 1914, clearing took two or three weeks, and if one was buying stocks across borders, the securities were sent one way and the money the other way by post.

Given all the modern databases and fast computers today, one might think clearing nowadays is instantaneous. But even now clearing can take days and even weeks. Slow clearing was at the heart of the GameStop and Robinhood crisis in January 2021.

Any disruption to the clearing process is costly. Suppose you sell Google stock. You are legally required to deliver the Google stock, and you have a legal right to receive cash at the same time — that is, the trade clears. If all you are doing is selling Google, a short-term disruption to the trading process might not be that problematic. However, most people have plans for the money they received from the sale. Perhaps you want to use it to buy Microsoft, and to pay for it you need the money from selling Google. If, for some reason, you don’t receive that money, you may default on the obligation to pay for Microsoft, even if you acted prudently and in good faith.

The 1914 crisis first affected the stock exchanges in Berlin, Paris and Vienna. To protect their local banks the authorities in those cities decided to suspend the clearing process, giving the dust time to settle. If everybody is trading locally that should not be a huge problem — at least nobody defaults because they cannot deliver cash. However, it is problematic if London remains open when Paris is closed. London-based banks send cash and securities to banks in Paris, but the reciprocation does not happen.

One might think this is an issue only for the sophisticated London banks doing business with the continent. Not so. The interconnectedness of the financial system meant that everybody was vulnerable. Even banks that thought they were sensible, like conservative rural banks trading only with major London banks and judiciously avoiding taking on too much risk, were nevertheless immediately affected.

First, those dealing with the continent got into difficulty, as they had to deliver to the continental trading partners but did not receive cash from their European counterparts.

Then their business partners got into trouble, and in short order those dealing with them faced difficulty. In the financial system everybody is exposed to everybody, whether they like it or not. We might think we are prudent by not doing business with those who like a lot of risk. However, if the people we are dealing with are exposed to risk, so are we.

The Crisis Goes Viral

A good example of the network effect, why financial crises are often said to be like a contagious disease, is as follows. Patient Zero infects the people around them, who in turn infect the people around them, and in short order everybody is sick. Similarly, the vulnerability quickly spreads from Bank Zero to all the other banks. The financial system is vulnerable not only because of risk but also — and even more so — because of how everybody is connected to everybody else.

London was at the center of the network in 1914, a place it had by then already occupied for a century and a half after eclipsing Amsterdam. The players change, but the network is always there, the source of simultaneous danger and prosperity. The network makes the economy efficient, creating wealth, good jobs and ample cheap goods, but it also transmits crises, both financial and medical. The plague in the 1300s was transmitted by trading networks, just as Covid-19 was in 2020.

If the 1914 crisis had been purely financial, nobody would really have cared besides a handful of people involved with the financial system. But it never is. The financial system directs resources for the real economy, and if it is not doing its job companies can’t borrow and invest. They can’t pay suppliers and salaries, the very essence of a systemic crisis. Economic activity grinds to a halt, and everybody suffers. That is the reason finance is so heavily scrutinized and regulated, and why the banks always get these obnoxious bailouts when they misbehave.

Once under way, the 1914 crisis followed predictable patterns. Market participants became very cautious, wanting the safest asset: gold. Banks deposited their excess funds to the safety of the Bank of England instead of lending them out, so that the crisis quickly spread from the City of London to the real economy as commercial credit evaporated.

In the beginning, the Bank of England and the leading financial institutions in the City of London tried to sort it out by themselves. However, things got worse, and it ultimately became the Treasury that made the most critical decisions. It is always like that. A serious crisis response needs the legitimacy of the political leadership — something an institution like the Bank of England, led by non-elected bureaucrats, simply cannot deliver.

Finally, both London’s and New York’s stock exchanges closed on July 31 and remained closed until January 1915. Why would one want to close the stock exchange? To prevent a vicious “mark-to-market” feedback loop. A company is solvent if the value of its assets exceeds the value of its debt. If the price of its assets falls, the company may find itself in difficulty or even insolvency because debt does not fall along with the value of the assets. To protect itself, the company may then sell its holdings.

That, however, makes prices fall even more, putting even more companies into difficulty, who will then also sell, and a vicious feedback loop is born. Soon everybody seems to be moving in lockstep, doing exactly the same thing, all selling and all buying the same assets. The resulting extreme volatility is the outward manifestation of a crisis in the making.

When the stock exchange closed, prices froze and the feedback loop was broken. Provided confidence has recovered by the time the market reopens, a crisis is averted. The same thinking is behind the “circuit breakers” on modern exchanges, which, if prices fall too much, halt trading to give market participants time to make sense of what is happening. The Chinese wanted to prevent the worst when they closed the Shanghai market on January 24, 2020, for six trading days when its Covid- 19 crisis broke out. The closing might have prevented the appearance of excessive volatility, but the market still opened down 8 percent.

The 1914 crisis showed that it does not take all that much for panic to happen when confidence evaporates. The very same efficiency that was so beneficial before a shock is a curse after it amplifies the initial shock into something much worse. Even rumors that a large institution might fail could cause panic. In 1914 it was clear that some banks were heavily exposed to cross-border lending and would fail in case of war. That was enough to trigger the crisis. People did not need the actual event of a war to make this happen; the mere expectation of one was sufficient.

The financial authorities in 1914 did the right thing, effecting a massive intervention, saving the City of London from the worst and, along with it, the real economy. By comparison, events in the 2008 and Covid-19 crises were quite mild. However, the ultimate consequence for the City of London was severe. It was the most important financial center in the world in 1914 but by 1918 was eclipsed by New York, which still retains the crown.

There are two main ingredients in the typical crisis story: excessive leverage and an interconnected financial system — in other words, too much risk and a dangerous network structure. But we don’t need extreme risk-taking for a crisis to happen. The 1914 crisis did not happen because of too much risk. No, instead, the forces of instability preyed on the very sophisticated financial system we depend on to give us efficient financial intermediation and leverage, helping the economy grow — but only in good times. That fantastic financial system also acts as the catalyst for financial crises. And that means it is all about politics.

It’s the Politics, Stupid

When Bill Clinton was running for president in 1992, his campaign’s unofficial theme was “It’s the economy, stupid.” Indeed, the main reason he defeated George Bush Sr. was the economic recession of the previous year. Now, it is the other way around — when explaining the market, one might say, “It’s the politics, stupid.”

Today’s primary source of financial risk is not excessive risk-taking; it is quantitative easing and low interest rates, Brexit and Donald Trump and the South China Sea, Russia, Ukraine, Qatar and Italy. The most severe financial crisis of the world today, that in Venezuela, is caused by politics.

Politics strongly shape the reactions to Covid-19. Government policies drive the financial markets. The financial markets react to central banks’ interest rate decisions and quantitative easing. When Covid-19 hit in early 2020 and the financial markets came under stress, the central banks were expected to step up to the plate and save the market. Not surprisingly, many commentators question why we simultaneously permit all the excesses in the financial system and also allow it to exercise the “Greenspan put” every time something bad happens.

The moniker “Greenspan put” is named after the former chairman of the Federal Reserve, Alan Greenspan, because every time the financial markets got the hiccups, the Federal Reserve was ready with a liquidity injection. If the Fed bails us out there is no reason to behave reasonably: profits are private, while losses are socialized. The consequence was that almost everybody was blind to the hidden risk being created before 2008 — those times were even labeled the Great Moderation. The socialization of losses and privatization of profits continue to anger activists on the left and the right and is one of the main drivers of populism.

Governments can be a direct trigger of systemic risk, as in 1914. The leading cause of the European sovereign debt crisis in the 2010s was not bank misbehavior but government profligacy.

The sub-prime mortgages that were so instrumental in the 2008 crisis were created because successive U.S. governments wanted to encourage homeownership for its poorest and riskiest citizens. However, the link between today’s politics and the instability of tomorrow is not clear except to a handful of people who closely monitor the relationship between the two. The very government policies that culminated in instability may have been launched with the best intentions, and their impact on systemic risk is often indirect and counterintuitive.

As a consequence, some commentators have made robust remarks on the origins of systemic risk. The late Fischer Black, of Black- Scholes formula fame, stated:

When you hear the government talking about systemic risk, hold on to your wallet! It means that they want you to pay more taxes for more regulations, which are likely to create systemic risk by interfering with private contracting. In sum, when you think about systemic risks, you’ll be close to the truth if you think of the government as causing them rather than protecting us from them.

Do policymakers recognize that governments also cause systemic risk? No. A few years ago I was at a large central bank conference in a European capital discussing how the European authorities should respond to the European sovereign debt crisis, then raging. In the closing panel, leading policymakers debated what should be measured, controlled and eliminated, all the standard things like excessive risk-taking, opaque financial instruments and perverse incentives. In the Q&A session, I asked the panel whether it was possible that the governments also caused systemic risk. The chairman of the panel shouted at me, “No, the governments are the solution, not the problem.”

Systemic crises are serious. Hundreds of millions of people suffered in the Great Depression, driving the politics that caused World War II. People lost their jobs, their savings and even their lives. Populism thrived. No wonder we would like to prevent them.

The first lecture of my London School of Economics course, “Global Financial Systems,” is about systemic risk. About 20 minutes into it I ask the students the following question: “Those of you who would want to live in a country without any systemic crises, please raise your hands.” In a typical year about 80 out of 100 students will do so. If I reverse the question and ask them if they want to live in a country with systemic crises, nobody raises their hands. If we were to ask most politicians, journalists, regulators and pundits the same question, I suspect most would agree with the students.

It is straightforward to prevent systemic crises: both Cuba and North Korea manage it quite well. Get rid of the financial system and voilà, no more systemic crises. Easy, but the costs are unacceptable. Who wants to live in Cuba or North Korea?

We need the financial system to take risk. With risk comes failure, an essential part of a healthy economy; an economy without failures does not take enough risk and does not grow fast enough.

While it is easy to regulate risk out of the financial system, it is not easy to do so without killing economic growth at the same time. On the other hand, too many bankruptcies and crises are also a sign that something is wrong. The financial system could be too unstable, with excessive risk, perhaps too much corruption, or money channeled to unproductive uses for political reasons. There is a balance between having too many and too few crises. We need to encourage the financial industry to take enough risk so the economy grows while also preventing too much risk from causing systemic crises. A classic riskreturn tradeoff, and one that is not trivial.

Economic crises afflict poor, less developed countries more frequently and severely than their rich and more developed counterparts, not least because developing countries are more likely to suffer from bad economic management and volatile politics. They depend on a small number of exports, often commodities, and are hence vulnerable to the whims of the global markets. It is hard to raise money in these countries, as the wealthy export their capital, forcing governments and other borrowers to raise funds in foreign currency from global financial centers. No surprise that sovereign debt, foreign exchange, and general economic crises are endemic in less developed countries.

It is not so with systemic risk: the very lack of development protects. Countries with underdeveloped financial systems, where credit cards and bank accounts are rare and everybody uses cash, are mostly safe from systemic risk. If the banks in such a country fail, most people can shrug it off. If my bank in London fails today, my debit card stops working, so I can no longer buy lunch. All my money is tied up in my bank accounts, and if I no longer have access to them there is nothing I can buy. Not only will I starve to death, but all the vendors selling to me will also not get any business.

In developed countries the safety of the banks is of primary importance. The more sophisticated the financial system becomes, the more vulnerable it is to disruptions. Systemic risk is the disease of the affluent.

We Had Forgotten History by 2008

Everybody thought we had solved the problem of crises by 2007. Why? Two reasons. The first is that systemic crises are quite rare, so nobody was worried. If you haven’t experienced one in your lifetime, you will not expect one to be around the corner either. The second reason is hubris. The experts had come to believe that modern risk management systems and regulations were sophisticated and robust. Policymakers and academics and practitioners all thought crises were associated with less developed economies, countries with weak financial regulations and plenty of corruption and abuse. The developed Western economies had found the magical way of preventing crises by combining efficient financial regulations and atomistic risk management. Hubris.

Atomistic risk management means that if we use sophisticated risk management techniques to manage all the small risks, they will not grow into big risks — voilà, crises are eliminated. That logic is fatally flawed. It is true that the whole of the financial system is made up of everything that’s going on within it, all the way down to individual transactions. We can know all about the atoms that make up the human body and be experts in biology, but all that knowledge tells us nothing about what makes a person tick. It’s the same in finance. We simply don’t know how to add up all the individual micro risks to assess the risk in a portfolio, bank or the system. If we think we do, all we end up with is a false sense of safety.

We thought risk had magically disappeared by 2007 because all the micro risks were under control. What we missed was that all the sophisticated risk management techniques had accomplished was to increase systemic risk. The reason is simple. The financial system is, for all practical purposes, infinitely complex, so no matter how intensively we study the system and how hard we try to control it, we still can focus only on a tiny part of it — systemic risk emerges precisely where we are not looking. Paradoxically, the sophisticated risk management techniques increased the complexity of the financial system, thereby creating new avenues for crises to emerge.

Take U.S. subprime mortgages, an important contributor to the 2008 crisis. The value of American subprime mortgages was about $1.3 trillion in March 2007. In normal conditions only a fraction of those borrowers were likely to default. In a disaster scenario, if half the borrowers defaulted and only half of the creditors’ money was recovered, total losses would be $325 billion.

While that figure sounds like a lot, it is quite small compared to the overall size of the American financial market. Right before the crisis in 2007, the outstanding value of bonds in the United States was $32 trillion, or twice the GDP, while the total value of the stock market was $20 trillion, so a $325 billion loss is about 1.6 percent of the stock market. A loss of that magnitude on the Standard & Poor’s 500 index has happened on 1,274 days since 1929, or 5.5 percent of the time. The actual subprime losses turned out to be much smaller.

How can a worst-case potential subprime loss of $325 billion cause so much damage when the stock market can suffer losses of $325 billion multiple times without batting an eye? The reason is that stock market losses are visible and expected. The mortgage losses were opaque; the risk was hidden, and it caught almost everybody by surprise. I say almost because some did anticipate it, like the heroes of Michael Lewis’s book The Big Short.

The crisis of 2008 was not the first time a financial crisis played out this way. Financial crises tend to be quite similar to one another, and the crises of 1763 and 1914 have plenty in common with 2008. Why, then, did 2008 happen? Hubris and forgotten history. While the policymakers of 100 years ago were alive to the danger of systemic risk, their 21st-century counterparts assumed the problem away.

I recall being in central bank conferences in the early 2000s and hearing most speakers argue that the central banks should focus only on inflation. Financial stability was not something the central banks should be concerned with because it is impure, sullying the pristine reputation of monetary policy, so crucial for price stability. Precisely why the financial authorities were caught by complete surprise in 2008.

The Bank of England, deciding in the early 2000s that all that mattered to it was monetary policy, proceeded to close down divisions focusing on financial stability and regulations — so when the crisis in 2008 started, it was desperately short of expertise. The then-governor of the Bank of England, Mervyn King, said in August 2007 after the crisis was already under way and Northern Rock was failing, “Our banking system is much more resilient than in the past. The growth of securitization has reduced that fragility significantly.”

It wasn’t only the central banks. Just about the only academic institution concerned with systemic risk was the London School of Economics. Credit goes to Charles Goodhart, who constantly reminded us that systemic risk was important and worth studying. Still, it was not good from a career perspective. I recall getting a referee report from a top journal on one of my papers on crises in 2003, rejected as “irrelevant because the problem of crises has been solved.”

The reaction to the Covid virus has its origins in the fumbling response to the crisis in 2008. Having been caught short in 2008, the policy authorities in 2020 responded with vigor, aided by strong political support and determined to prevent a repeat of 2008. Policymaking is often like that. Underreaction followed by a rearview-mirror-guided overreaction. Unfortunately, though, crises are all fundamentally the same. The details differ, and regulations target the details much more than the fundamentals.

Covid-19 offers a fantastic example of the trade-off between safety and growth: do we lock the economy down to prevent the virus from spreading but at the expense of very significant economic damage, or keep everything open and hope for herd immunity? The answer to that question shows why we need political leadership; the authorities concerned with health tend to prefer lockdown, and the economic authorities to keep things open. It is the job of the prime minister or president to arbitrate and decide on the best course of action. Directing the fight against Covid-19 is not a job that can be delegated to officials.

We have the same debate in the financial system. Do we seek risk and so deregulate the system, hoping for more growth, or do we nail the system down to prevent crises? Both camps have their adherents, and it is the job of the political leadership to decide.

The Covid-19 and 2008 crises have a lot in common. Like most, they comprise a fourstep process. Willful dismissal of the threats before the crisis event, followed by a weak initial reaction. When things get so bad they can’t be ignored, we get overreaction and eventually, as we gain knowledge and experience, an uneasy balance between safety and growth. But the differences between the two crises are more important.

When the Covid-19 crisis was under way, I wrote a piece with three of my LSE colleagues, Robert Macrae, Dimitri Vayanos and Jean- Pierre Zigrand, titled “The Coronavirus Crisis Is No 2008,” in which we argued that these two crises were quite different. The 2008 crisis originated within the financial system, caused by its willingness to take risk, aided by the willful ignorance of the dangers and excessive complexity. Covid-19 came from nowhere, certainly not the financial system. While the policy authorities reacted strongly to Covid-19, finance was a small part of that, and we haven’t seen any financial crisis come out of Covid-19.

So Covid-19 impacted the financial markets, but it cannot be called systemic in a financial sense. What Covid-19 does do for us is to provide a handy framework for thinking about financial regulations, what works and what doesn’t work.

the necessary scaffolding to make the argument. But as a brief preview, recent empirical evidence suggests that how the financial authorities reacted to Covid-19 has sharply increased moral hazard — that is, the financial markets’ expectation that the central banks are ready, willing and able to step in and bail them out. They, as always, will react by taking yet more risk.

Who Is To Blame? Banksters? Politicians? Regulators? Nobody?

When a systemic crisis happens, someone must be to blame. Is it the “banksters” who took too much risk and abused the financial system? Is it the regulators who were asleep on the watch? The politicians who cheered the banks on and protected them from scrutiny? The households that borrowed too much and saved too little? Someone made the decisions that led to the crisis. Should they be put in jail?

Take the example of Spain, going through a huge boom after it joined the euro, fueled by the influx of very cheap money. While some banks were careful, others were not — especially the savings banks, the cajas. The lending boom seemed beyond anyone’s control from about 2005, and everybody enjoyed the party. Who is responsible? The designers of the euro? The European Central Bank? The Spanish government that joined the euro? The international banks that lent to Spain? The central bank of Spain? The Spanish commercial banks? The borrowers?

If we try hard enough, we can always find someone to blame. As Cardinal Richelieu said, “Give me six lines written by the most honest man, and I will find something there to hang him.”

Still, it is not that easy to find someone to blame if we can’t use the methods of Cardinal Richelieu. My native country, Iceland, was the country worst hit by the global crisis in the autumn of 2008, and since then has been trying very hard to find someone to blame. The government convened a special court to prosecute the then-prime minister for allowing the crisis to happen, eventually convicting him of negligence — to my mind, a miscarriage of justice. If we were able to prosecute politicians for incompetence, there wouldn’t be many left.

The Icelanders had better luck in going after the bankers, some of whom have spent years in jail. But it took an extraordinary effort, a special prosecution service, and financial police with considerable resources and very strong political backing, all helped by questionable actions like long solitary confinement for those accused but not yet convicted. In the end, the bankers were convicted not for causing the crisis but for minor misconduct relating to individual transactions — just like the U.S. authorities got Al Capone for tax evasion, not for being a mafia boss.

The regulators and most of the political leadership got off scot-free, even if they are as much at fault as the bankers. The Central Bank of Iceland, which bore so much blame for allowing the crisis to happen, fired the governors but promoted everybody else. So, ironically, the career bureaucrats who were responsible benefited.

The Icelandic case is an extreme form of what we have seen in other countries. The prosecutors can go after bankers for specific misconduct, some abuse of office, but they cannot convict anyone for causing a crisis. Spain sent its former finance minister and IMF chief Rodrigo Rato to jail for four years and six months for embezzlement. He was not convicted for the failure of his bank (Bankia), rescued at huge expense by the Spanish taxpayer. He was not punished even for the losses suffered by 200,000 savers whom he had persuaded to buy subordinated Bankia bonds right before it failed. No, he misused his corporate credit cards.

One might wonder what the Spanish regulator was doing, not only condoning but also actively encouraging such abuse of unsophisticated investors. Why aren’t they prosecuted?

The former CEO of Barclays, John Varley, is the only CEO of a global bank to face charges because of his conduct in the financial crisis. However, that was only because of how he tried to save his bank, not for getting it in trouble in the first place. The United States has sent at least 35 bankers to jail for crimes related to the financial crisis, most relating to small amounts of money at small banks for personal gain. One person, known as Fabulous Fab — real name Fabrice Tourre — was convicted of crimes relating to structured credit products while working as a junior employee at Goldman Sachs. He and everybody else convicted in the United States were small fry.

Even when there is clear abuse, we cannot find anyone to blame. The best case is LIBOR [London Interbank Offered Rate], the standard benchmark interest rate used to decide on interest rates charged on mortgages and loans worldwide. It was surprisingly easy to manipulate LIBOR, as it was based on an average of banks’ estimates of market interest rates, and the employees making the submissions in each bank had some leeway in the number they came up with.

Is it 5.25 percent or 5.26 percent? A derivative trader who knows which of these two numbers is to be submitted has more than an even chance of profiting, and it is alleged that employees across banks colluded in their submissions, guaranteeing profits. The banks maintain it was all the fault of rogue junior employees, and nobody higher up had an inkling of the abuse. Still, they didn’t question the profits from manipulating LIBOR.

Did the regulators know? While strenuously denied, many observers allege that banks deliberately posted excessively low numbers during the 2008 crisis with the acquiescence of the regulators in order to lower funding costs for banks in difficulty.

The manipulation of LIBOR was very costly for lenders and borrowers, and one might think the abuse would be severely punished. Not so. Employees in several banks were found to have been manipulating LIBOR, and some banks have admitted to doing exactly that. Just one person has, at the time of writing, been punished: a junior UBS and then-Citigroup employee who was sent to jail by the British authorities for 11 years. No senior manager, financial institution, or regulator has been prosecuted or convicted for the LIBOR abuse.

The reason is that it is really difficult to find anyone guilty in a court of law. The prosecutors in the United Kingdom and the United States have tried but mostly failed. The bankers might be responsible for causing a crisis in the court of public opinion, but the Icelandic bankers, like their counterparts in other countries, were careful not to break the law.

Stupidity is not a crime. Greed is not a crime. Legally manipulating the rules is not a crime. Recklessness was not a crime in 2008, even though it has become a criminal offense in the United Kingdom since.

The bankers, regulators, and politicians cannot be convicted of a crime that does not exist, and we cannot (or should not) change the law retroactively. When we do a postmortem on financial crises, we see that the bankers excessively expanded their banks, taking on too much risk, not properly scrutinizing lending decisions and ignoring liquidity risk. The regulators looked the other way, and the politicians cheered the excesses. It may be greedy, incompetent, arrogant and immoral, but not, so long as the letter of the law is complied with, illegal.

All we have is specific misconduct, and even then it is really hard to prove. It involves complicated financial transactions understood only by a handful of experts and difficult to explain in a court of law to non-expert judges and juries. Especially since conviction hinges on a fine interpretation of rules, the burden of proof is on the prosecutors, and it is not easy to find people guilty beyond a reasonable doubt.

There is another, more insidious problem in finding banks or bankers guilty, and that is the problem of too big to fail, illustrated by an example from the world’s second largest bank, HSBC. In the years before 2010, it neglected to monitor transactions involving drug traffickers in its Mexican subsidiary. Because these transactions were in U.S. dollars, they were subject to clearing in New York and hence to U.S. law.

The U.S. government refused to prosecute HSBC because, as Eric Holder, the U.S. attorney general at the time, explained, “I am concerned that the size of some of these [financial] institutions becomes so large that it does become difficult for us to prosecute them.” In other words, if the U.S. government had decided to go after HSBC it might have triggered its failure, and, because HSBC is a systemically important financial institution, such failure might result in a systemic crisis. The largest banks have a get-out-of-jail-free card.

What the U.S. authorities ended up doing is to fine the banks, and in the aftermath of the 2008 crisis, banks have paid over $320 billion in fines without admitting to any misconduct. The fines have become yet another revenue source for the government, a tax by another name borne by the banks’ clients in the form of higher fees. Other countries haven’t even gone that far.

Systemic risk is the chance that the financial system does not do its job, perhaps that a major financial crisis causes an economic recession. It happens because of the typical dilemma ever-present in the financial system. We want economic growth, and that necessitates risk. With risk comes the chance of failure and crisis. It is hard to prevent systemic risk because it emerges in the most obscure parts of the financial system, exploiting unknown vulnerabilities, making the policymakers’ job difficult — but certainly not impossible. We know quite a lot about why crises occur and how to prevent them. Yet they are all too familiar.

And the economist on the spine is ...

no doubt raj chetty, the India-born American economist on the spine of the 2022 editions of the Review, checks all the right boxes and then some: PhD from Harvard … tenure at Harvard at the ripe young age of 28 … John Bates Clark Medal for the most distinguished economist under 40 … a “genius grant” from the MacArthur Foundation … Best of all, he actually earned all the hype with pathbreaking research using big data to measure economic and social mobility in America. Actually, his most striking (and easily accessible) work is through Opportunity Insights, his nonprofit research organization that uses vast troves of public and private data to measure social capital (economic connectedness, cohesiveness and civic engagement) right down to the zip code level. If you are just curious or want to get your hands dirty with big data, check out OpportunityInsights.org.

— Peter Passell