• … [this book] originated, strange to say, as a study of the history of the international bond market. I came to realize in the course of my research, however, that the bond between creditor and debtor was only one of many bonds I needed to consider; and that in many ways the bond market was interesting precisely because it concerned itself with these other bonds as well: above all, the usually implicit contractual bonds between the ruler and ruled, the elected and the electors, but also bonds—more often (though not always) contractual—between states.
  • ——Niall Ferguson, The Cash Nexus: Money and Power in the Modern World, 1700-20001 p. 20
Niall Ferguson’s The Cash Nexus, published in 2001, convinced me that the way that economists approach central banking, monetary policy, and financial regulation was incomplete and misleading. I came away from the book thinking that all elements of financial policy are geared toward the goal of enabling the government to allocate credit to its preferred uses, especially its own spending.

The focus of government on credit allocation receives no attention in economics textbooks. Ignoring this fundamental purpose of financial policy, economists act as if central banks exist solely to conduct monetary policy for stabilization purpose. The textbooks, if they do not completely overlook financial regulation, mention it in a hand-wavy way as helping to keep the financial system sound.

In an alternate history of economic thought, the paradigm that emerges from The Cash Nexus would have kicked off a substantial research program in financial policy. This research program might have been sufficiently fruitful that the 2022 Nobel Prize in economics would have gone to its leading scholars and perhaps Niall Ferguson as well. Instead, that honor went to Ben Bernanke, Douglas Diamond, and Philip Dybvig, whose insights I regard as pedestrian in comparison.

But in fact, there was no such follow-on research program. Ferguson’s framework was never filled in. Had economists absorbed Ferguson’s approach to thinking about financial policy, the financial crisis of 2008 might not have caught them with their pants down. Also, we might have arrived at a different interpretation of the policy response, especially the “quantitative easing” undertaken by the Fed. We might have a better understanding of the strengths and weaknesses of our present financial situation.

A Product of Its Time

The Cash Nexus takes a historical and institutional view of the development of government finance and central banking. Even as I find its approach relevant for understanding government involvement in financial markets today, the current context is quite different from what Ferguson would have foreseen. To understand why, I think it helps to see the book as a product of its time.

Although its copyright date is 2001, The Cash Nexus evidently was put to bed early in 2000. In chapter 10, “Bubble and Busts: Stock Markets in the Long Run,” the latest data point given is from February 2000. The book makes no mention of the market meltdown that got underway with the collapse of the “dotcom” stocks, beginning in March of 2000. This crash would have been a very relevant event to include in the book had the publication process allowed for it.

Other events that took place after the book’s publication include:

  • • the terrorist attacks of 9/11/2001
  • • the invasions of Afghanistan and Iraq
  • • the widespread acceptance of the euro (it was launched in 1999, but it was regarded as a bold experiment, not necessarily destined to succeed. Ferguson refers to it as “the EMU,” for European Monetary Unit, which was how this novel currency was known at the time.)
  • • the financial crisis of 2008 and the response to that crisis, including bank bailouts, the Dodd-Frank legislation, and “quantitative easing.”
  • • the protest movements of Occupy Wall Street and the Tea Party
  • • the populist shocks in 2016 of Donald Trump’s election and the vote for Brexit
  • • the pandemic starting in early 2020 and the fiscal and monetary response
  • • Russia’s invasion of Ukraine and the imposition of economic sanctions by the West in response
  • • movements by the Trump and Biden Administrations in the direction of “decoupling” the American economy from China
  • • a surge in inflation in 2022, leading the Fed to sharply raise interest rates

Financial considerations, and what Ferguson calls “the usually implicit contractual bonds between the ruler and the ruled,” were heavily implicated in all these events. But an environment of adverse shocks and reactions against globalization were not what was foreseen in 1999.

Instead, 1999 might have been the peak year for the perceived triumph of capitalist democracy, which was how many people interpreted Francis Fukuyama’s End of History, published in 1992. Tony Blair, who became Britain’s Prime Minister in 1997, and Bill Clinton, who became President in 1993, represented the triumph of the elite centrist approach to political economy. Economists were praising central bank independence as the antidote to inflation. They saw a solution for economic development embodied in the “Washington consensus” of free trade, deregulation, and democratic institutions.

“As of 1999, prosperity, globalization, movements toward democracy, and elite centrist politics were the norm. In that context, what Ferguson called ‘the square of power’ (tax bureaucracy, legislature, government debt, central bank) apparently had arrived at a benign equilibrium.”

As of 1999, prosperity, globalization, movements toward democracy, and elite centrist politics were the norm. In that context, what Ferguson called “the square of power” (tax bureaucracy, legislature, government debt, central bank) apparently had arrived at a benign equilibrium. If The Cash Nexus did not include a clarion call to undertake new research on how governments could establish a sound financial footing, that might have been because of this sense of having reached an end state. The interesting problems appeared to be solved.

During the Clinton Administration, the United States seemed capable of becoming the world’s policeman. Ferguson wrote,

  • Under President Clinton, the aims of American foreign policy were extended beyond the defense of allied states—the number of which has increased as a result of NATO enlargement—to include the termination of civil wars in a number of politically sensitive regions, and the occasional use of military force to protect the rights of persecuted minorities in certain countries. p. 394.

Although George W. Bush campaigned against this interventionist approach, and he seemed inclined to focus on domestic policy, his Presidency turned out otherwise. He will instead be remembered as the President who, in the aftermath of 9/11, initiated the invasions of Afghanistan and Iraq.

President Clinton had allowed his more progressive economic proposals to be vetoed by financial markets, because of their potential adverse consequence for the interest rate on government debt. This provoked Clinton’s adviser James Carville to complain that if he could be reincarnated into a position of power, he would want to come back as the bond market.

From World War Two until 1980, the ratio of federal government debt to GDP declined from about 100 percent to just 30 percent. This was largely because inflation and growth in real GDP diluted the debt/GDP ratio, even though the budget was in deficit most years.

It was under President Clinton that the government budget of the United States last ran a surplus. Ferguson wrote that “The Clinton surpluses of the late 1990s have raised the prospect of substantial if not total repayment of the federal debt.” (p. 127) Ferguson was far from the only observer contemplating a future with a debt/GDP ratio of zero or less. During the debates over tax-cut proposals made by President Bush in 2001, Federal Reserve Chairman Alan Greenspan chimed in with support for the proposals. His argument was that without federal debt, monetary policy would be impossible to conduct, because it consists of buying or selling government bonds to increase or decrease the money supply, respectively. Ergo, tax cuts were needed in order to prevent the disappearance of government debt. In hindsight, this ridiculous argument came to be refuted, as under Greenspan’s successor the Fed showed that it could purchase mortgage securities just as readily as government bonds. Furthermore, no one need to have been concerned about the government paying off its debt. By now, the debt/GDP ratio has returned to World War Two levels.

The twenty-first century would see new norms for conducting fiscal policy. In January 2008, Larry Summers spoke for most economists and policy makers of his generation when he said that fiscal stimulus should be “timely, targeted, and temporary.” Instead, we had fiscal blowouts in response to both the financial crisis of 2008 and the pandemic of 2020. Moreover, once those crises had passed, there seemed to be no inclination to shrink the debt tumor.

The War-Making Machine

Ferguson argued that throughout history states have periodically engaged in war. War requires resources. Military innovation has made war increasingly capital intensive. Therefore, the states that survived were those that were good at mobilizing resources in times of war. This meant the creation of institutions for raising funds.

In order to mobilize resources, the government must be able to collect taxes. Ferguson wrote that this is best undertaken with a tax-collection bureaucracy.

Tax collection requires compliance. Rulers found that compliance was easier if taxes were approved by an assembly representing constituents. In medieval times, the assemblies were composed of feudal lords. In more modern economies, the range of represented constituencies expanded, and along with it the range of those eligible to participate in electing representatives. On this view, countries became increasingly democratic in order to enable rulers to collect more taxes.

Times of war require more revenue than times of peace. Governments that could borrow during wartime and repay debt afterward could be more successful in war. Ferguson argues that government’s ability to issue securities became an important factor affecting a state’s ability to wage war. Successful governments were able to come up with debt instruments and commitment mechanisms that were attractive to investors.

In order to borrow during times of war, government needs large financial institutions that can underwrite its securities. Ultimately, states arrived at the idea of a central bank to ensure the stable functioning of a market for government debt. But states still needed large private underwriters. In the United States, these were known as the “primary dealers,” which purchased securities issued by the Treasury and through which the Federal Reserve Bank of New York conducted open market operations.

As Ferguson pointed out, the institutions that emerged to finance war evolved to achieve other ends. Representative assemblies made states more responsive to the desires of the public, leading to the creation of the welfare state. The proportion of tax revenue dedicated to social insurance steadily increased.

Over the past century, the public has come to expect government to manage the overall state of the economy. Peacetime government borrowing increased during times of recession. Central banks were tasked with raising the money supply enough to mitigate recessions without raising it too quickly to stoke inflation. Financial regulations and government backing of financial institutions were introduced in order to try to make saving and investing more secure for consumers. Recently, governments went on a spending binge to “combat the pandemic” as if they were fighting a war.

When the state began to backstop financial institutions, by providing deposit insurance for example, this gave government the ability to influence the direction of investment. Regulations are used to steer capital toward purposes favored by government. Mortgage lending was an important example, and in the United States taxpayers were put at risk.

The Savings & Loan Crisis

As part of the New Deal, the government created the Federal Housing Administration (FHA), the Federal National Mortgage Association (Fannie Mae), and the Federal Home Loan Bank Board (FHLBB). The latter did not lend directly to consumers, but it provided backing to the savings and loan industry, which became the main source of mortgage funds. FHA, Fannie Mae, and the S&Ls were encouraged to offer 30-year, fixed-rate, level-payment, amortizing mortgages, in lieu of the short-term balloon mortgages that suffered widespread defaults in the 1920s and in the Great Depression.

By the 1970s, the 30-year fixed-rate mortgage itself caused distress, as inflation and interest rates rose. Households were financially secure, but lenders found that their cost of funds rose while the income from outstanding mortgages stayed constant. Had their balance sheets carried assets at current market values, most S&Ls would have been bankrupt by the end of the 1970s.

Government officials did their best to keep troubled institutions going. Fannie Mae, although technically bankrupt by 1982, was kept alive long enough to recover when interest rates finally declined in the latter part of the 1980s. Even though the market value of older mortgage loans held by S&Ls was barely more than half their book value, FHLBB allowed S&Ls to report as assets on their balance sheets the full book value of those loans. But losses at the S&Ls worsened, in spite of attempts by the FHLBB to keep them afloat. These attempts included the creation of the Federal Home Loan Mortgage Corporation (Freddie Mac), which helped bankrupt S&Ls maintain fictional solvency by exchanging securities for loans in portfolio without the FHLBB requiring the S&Ls to mark to market the securities on their books, which were trading at prices far below the book values used by the S&Ls. The longer that the policy of “extend and pretend” was followed, the deeper in the hole the industry fell. The end result was that Congress had to vote substantial funds for a bailout to pay off depositors at the failed savings and loans.

The 2008 Financial Crisis

In reaction to the S&L crisis, regulators introduced capital requirements for depository institutions (primarily banks, now that the savings and loan industry had been decimated) that used risk weights. For the riskiest types of assets permitted, including mortgage loans, the requirement was that $8 in capital had to back each $100 in assets. But other assets—including mortgage securities issued by Freddie Mac and Fannie Mae—were given lower capital requirements, and government debt was assigned the lowest capital requirement of all. For capital efficiency purposes, banks now had a strong incentive to sell to Freddie and Fannie any mortgage deemed “investment quality” by those agencies. In effect, those two government-sponsored enterprises were used to allocate capital toward mortgage loans bundled into securities. For banks, commercial lending and “non-conforming” mortgage loans were relatively disfavored by the risk-based capital regulations.

In 2001, risk-based capital requirements were changed under what was called the Recourse Rule. Private investment banks, mostly on Wall Street, were given the ability to issue mortgage securities that could be given low risk weights if held by commercial banks. To do so, the securities had to be carved into “tranches” in which the portions that were the last to bear losses in case of default received AA or AAA bond ratings and were considered suitable for banks, while the remaining tranches that were the first to bear losses were held by other investors. Under the Recourse Rule, the low-risk tranches in these complex Collateralized Mortgage Obligations (CMOs) enabled banks to hold assets backed by mortgage loans that were not of investment quality (subprime loans), while enjoying the advantages of low capital requirements. Dealers that made markets in these low-risk securities were able to use them as collateral for repurchase agreements (repos), which are short-term loans that dealers use to finance their security inventories.

Expanded mortgage lending fueled a boom in house prices, which collapsed in 2007. By 2008, it became evident that the risk in CMOs was much higher than the bond rating agencies had calculated. This threatened the solvency of some banks that had large mortgage security portfolios. But the most devastating impact was on securities dealers. In the repo market, even highly-rated mortgage securities became unacceptable as collateral. Most of the major investment banks could not roll over their repo loans, and they sought mergers with commercial banks, with the encouragement of regulators. Previously, investment banking had been separated from commercial banking by the Depression-era Glass-Steagall Act, but in the crisis Glass-Steagall was buried. Even so, one investment bank, Lehman Brothers, was unable to come up with a rescue partner, and its bankruptcy brought the financial crisis to a boil.

The Treasury and the Fed determined that funds were needed for a large-scale financial bailout, which became known as the Troubled Asset Relief Program, or TARP. Officials sold this $800 billion bailout to the press and in turn to the public as a way to prevent a collapse of retail banking as had occurred in the 1930s. But I believe that most banks (and therefore most households) would have escaped unscathed without a bailout. Only a few of the large banks that participated heavily in the repo market were in danger.

The government’s real concern, I believe, was with the dealers in mortgage securities, the largest of which were also the primary dealers in the Treasury security market. Bailout funds that went into the financial sector (some of the taxpayer largesse went to other special interests, including auto manufacturers) mostly went to the banks that absorbed the primary dealers. The government was worried about the primary dealers in its own securities, not about households’ savings accounts.

After 2008, bank regulators and the Dodd-Frank financial reform act discouraged mortgage lending by imposing strict credit standards for mortgage origination. The effect was to end the favorable status of housing in capital allocation. Households found it much harder to qualify for mortgages. Housing starts remained depressed for more than a decade. A shortage of housing emerged, causing rents and prices to rise faster than overall inflation.

Much of the capital that did not go to finance housing went instead to finance large government deficits, especially during the pandemic years of 2020-2022. Some investment also went into other financial assets. In the decade between the financial crisis and the pandemic, there was a surge in stock prices, bond prices, valuations of firms funded by venture capital, and cryptocurrencies.

The True Meaning of Quantitative Easing

If the bank bailouts of 2008 have been widely misinterpreted, then so has another policy adopted in 2008, known as Quantitative Easing (QE). As I suggested above, the bank bailouts were undertaken to protect the primary dealers in government securities, not household deposits. Quantitative easing was explained as a way to expand the money supply at “the zero bound.” Instead, it should be interpreted as a form of debt management.

The “zero bound” story to explain QE is that once short-term interest rates approached zero, the Fed had to purchase other securities, including long-term treasuries and mortgage securities, in order to undertake further monetary expansion. But the actual way that QE was undertaken is not consistent with this story. Rather than use QE to expand bank lending and the money supply, the Fed introduced a policy of paying interest on reserves (IOR) in order to induce banks to hold onto reserves rather than lend them out. If the Fed had truly wanted the money supply to expand, it would not have introduced IOR. At the “zero bound,” IOR would have been zero.

Years later, the Fed expanded its balance sheet further with what it called reverse repurchase agreements. It was funding its portfolio the way that a securities dealer would, with short-term borrowing in the repo market. With IOR and reverse repo, the Fed was paying a short-term interest rate to fund purchases of long-term securities. The interest rate that it paid was always above zero. There was never any “zero bound.”

A better way to understand QE is to view it as debt management, overriding the Treasury. The Treasury would issue a mix of long-term debt and short-term debt. By buying the long-term debt and financing it with short-term borrowing, the Fed was converting the Treasury’s long-term debt into short-term debt.

Ultimately, the Treasury and the Fed are both agencies of the government, and we can view their actions in combination. Economist John Cochrane uses the metaphor that the government has two pockets. With its right pocket (Treasury), it issued some long-term debt. With its left pocket (the Fed) the government converted that long-term debt into short-term debt.

Until 2022, QE helped lower the government’s overall interest costs without having the Treasury withdraw from the long-term bond market altogether. Thus, QE helped to avoid atrophy in that market. But when interest rates rose in 2022, the government overall was stuck with more short-term debt and higher interest costs than it would have had without QE. The Fed in fact incurred large losses as the short-term interest rates that it had to pay for IOR and reverse repo rose above the rates it was receiving on its long-term Treasuries and mortgage securities.

I am suggesting that we view the TARP program in 2008 and the Fed’s subsequent QE through a Cash Nexus lens. The government’s top priority is to ensure that its ability to borrow is never interrupted. TARP kept the primary dealers intact; and QE kept the long-term debt market functioning, while allowing the government to take advantage of the lower cost of short-term debt—at least until the need to confront the outbreak of inflation in 2022 caused QE to backfire from a debt management perspective.

Recent Developments and the Square of Power

As of early 2023, a number of recent developments served to illustrate the ongoing relevance of Ferguson’s “square of power” model. Recall that the square includes the tax-collecting bureaucracy and note that prior to the 2022 election, the Democratic Congress approved President Biden’s request to appropriate funds for a big increase in hiring by the Internal Revenue Service. The next pillar is a representative legislature, and the Democrats’ loss of the House sets up conflicts with the Biden Administration over fiscal policy, and even the appropriation for more IRS agents is up for negotiation. As many countries stretch their borrowing to unprecedented levels, the third pillar of the government debt market is wobbly. In the United Kingdom, a revolt by bond investors vetoed Prime Minister Truss’ proposals for supply-side tax cuts, terminating her short-lived premiership. Finally, in America, the fourth pillar—the central bank—had spent months engaged in a delicate balancing act. On the one hand, interest rates had to rise dramatically in order to curb inflation. On the other hand, the rate increases had to be gradual and well telegraphed ahead of time in order to avoid catching primary dealers and other large institutions unprepared, which would have caused distress at those key institutions.

Another recent development was the collapse of many businesses associated with cryptocurrencies—most dramatically FTX and Alameda, two entities controlled by Sam Bankman-Fried who became notorious as an accused swindler. This development illustrates a major take-away from The Cash Nexus: among the “other bonds” in which the bond market is embedded is the bond between banks and government.

The government needs banks in order to ensure that credit is available for its own spending and other favored uses. And banks need government in order to be able to perform their function of holding risky, long-term assets and issuing riskless, short-term liabilities so that households and businesses can do the opposite: hold riskless, short-term assets (like checking account deposits) while issuing risky, long-term liabilities (like mortgages). Government enables banks to do this through a combination of enforcement of debt contracts, explicit backing of financial institutions (deposit insurance, borrowing privileges such as the Fed’s Discount Window), implicit backing (“too big to fail”), and regulatory enforcement intended to reassure the public that they are protected against fraud or reckless behavior.

The cryptocurrency sector lacked government protection. Many crypto enthusiasts do not want it, because they would prefer not to be corrupted by or beholden to governments. But the chaos that hit the market in 2022 makes the prospects for a separation of finance from state seem remote. Readers of The Cash Nexus would not be surprised.

For more on these topics, see

The Cash Nexus bids us to consider the linkage between finance and government. Both banks and government must be perceived as long-lasting in order to function. You might have no problem eating at a restaurant or buying at a clothing store that may have to go out of business next month. But you would not take your banking business to an institution that is at high risk of failure. And you would not feel a need to obey a government that is likely to be overthrown soon. Support from financial institutions helps governments convince citizens of their staying power. And support from government helps financial institutions convince citizens to trust them with their savings.


[1] Niall Ferguson, The Cash Nexus: Money and Power in the Modern World, 1700-2000. Paperback. Basic Books, 2002.

The hardcover was published in 2001 under the title The Cash Nexus: Economics and Politics from the Age of Warfare Through the Age of Welfare.

*Arnold Kling has a Ph.D. in economics from the Massachusetts Institute of Technology. He is the author of several books, including Crisis of Abundance: Rethinking How We Pay for Health Care; Invisible Wealth: The Hidden Story of How Markets Work; Unchecked and Unbalanced: How the Discrepancy Between Knowledge and Power Caused the Financial Crisis and Threatens Democracy; and Specialization and Trade: A Re-introduction to Economics. He contributed to EconLog from January 2003 through August 2012.

Read more of what Arnold Kling’s been reading. For more book reviews and articles by Arnold Kling, see the Archive.

As an Amazon Associate, Econlib earns from qualifying purchases.