The Monetary Rubicon and the End of Policy Orthodoxy

_{Photo by Michael Guesev, Doha, Qatar, 2025}

To better understand all of the content in this post I suggest first reading the following:

1. Overview on status of Dollar in the global economic financial system: https://therationalmargin.com/the-dollar-standard/
2. Overview on Fixed Income Mechanics: https://therationalmargin.com/fixed-income-investment/
3. Intro to market liquidity and Treasury actions: https://therationalmargin.com/us-equity-outlook/

Introduction: The Monetary Rubicon and the End of Policy Orthodoxy

For most of the twentieth century, central banking was a discipline of managing the price of money within a regime of reserve scarcity. Prior to the Global Financial Crisis (GFC) of 2008, central banks largely adjusted short-term interest rates (guided by paradigms like the Taylor Rule and the Phillips Curve) to influence economic activity. Their balance sheets were minimal and passive, dictated by the public’s demand for physical cash and small reserve requirements. Liquidity in the banking system was tight by design: the central bank supplied just enough reserves to keep overnight rates on target, relying on the interbank market to distribute that liquidity efficiently.

That orthodox model is now a relic. Over the past 25 years, the global financial architecture has been fundamentally reshaped by an unprecedented experiment in monetary policy known as Quantitative Easing (QE). No longer content to merely set the price of money, central banks began also targeting the quantity of money by expanding their balance sheets on a massive scale. From the Bank of Japan’s tentative forays in the early 2000s to the multi-trillion-dollar deluge of QE during the COVID-19 pandemic, central banks morphed from lenders of last resort into market makers of last resort and the primary managers of duration risk for debt markets.

This evolution represents a crossing of the monetary Rubicon. Central banks are no longer neutral referees on the sidelines; they are now the biggest players on the field. By late 2025, the consequences have evolved into a state of Financial Dominance, where the stability of sovereign debt markets and bank balance sheets effectively holds monetary policy hostage. No longer can central banks single-mindedly focus on inflation targets; their actions determine the solvency of pension funds, the liquidity of Treasury markets, and even the fiscal capacity of governments.

Defining Financial Dominance

Financial dominance is distinct from the classic concept of fiscal dominance. Under fiscal dominance, a central bank is effectively compelled to monetize government deficits (printing money to buy Treasury debt) to avert a sovereign default. Financial dominance, by contrast, is a broader and more sticky condition: the central bank becomes constrained by the imperative to maintain the solvency and functioning of the financial system itself. In a financially dominant regime, public and private debt levels are so high, and asset prices so sensitive to interest rates, that the central bank faces significant difficulty in raising rates aggressively to fight inflation without risking a cascade of bankruptcies, a banking crisis, or a bond market crash.

Put simply, financial stability constraints override the central bank’s inflation mandate. Balance-sheet losses, liquidity crunches, or solvency fears in the financial sector end up dictating policy decisions more than traditional macroeconomic indicators do. Interest-rate policy becomes subordinate to preserving debt sustainability and preventing capital destruction at systemically important banks. In quantitative terms, the point of no return into financial dominance is reached when raising rates triggers asset-price declines severe enough to threaten bank capital adequacy, forcing the central bank to reverse course back to easing, regardless of the inflation outlook.

The implications of this structural shift are profound, as evidenced by central bank behavior in recent years:

• U.S. Federal Reserve: After expanding its balance sheet to nearly $9 trillion, the Fed in late 2025 was compelled to halt quantitative tightening fearing that further shrinking of the balance sheet would push bank reserves below their lowest comfortable level of reserves (LCLOR), destabilizing overnight funding markets in the process. Preserving ample reserves took priority over the inflation mandate.
• Bank of England: The Bank of England’s attempt to unwind QE led to heavy mark-to-market losses on its bond portfolio, losses that the U.K. Treasury (ultimately the taxpayer) had to indemnify. This directly entangled monetary policy with fiscal finances, politicizing the central bank’s decisions and limiting its freedom to tighten policy.
• Bank of Japan: After decades of aggressive QE, the BOJ owns more than half of all Japanese government bonds. It cannot allow interest rates to rise meaningfully without imperiling domestic insurers and pension funds that suffer from large negative duration gaps (long-term liabilities far exceed the duration of their assets). The BOJ is essentially trapped into capping yields to prevent a solvency crisis: an extreme case of financial dominance verging on fiscal dominance.

To grasp this new reality, traditional textbook models (like the money multiplier or simple fractional reserve banking) are woefully inadequate. One must instead understand the hierarchy of money, the plumbing of funding markets (like repo), and the binding constraints imposed by post-2008 financial regulations. This work undertakes an exhaustive examination of how QE and related policies have structurally transformed modern finance into an excess-reserves system governed by bank capital rules and sovereign debt management. We demystify QE as a sterilized asset swap, a mechanism of duration extraction rather than money printing, and show how it compresses risk premia and forces portfolio rebalancing. We also explore advanced models of unconventional policy (such as the Wu-Xia shadow rate framework) that quantify the hidden easing provided when nominal rates hit zero.

The transition from the old regime to the new is characterized by several key structural shifts. Key differences between the Old and New Regimes include:

• Policy Objective: Shift from a focus on price and employment stability to an extended mandate that explicitly includes financial stability and market functioning commonly known as the Fed put.
• Operational Target: Shift from controlling a short-term interest rate (Fed funds) to also managing the quantity of reserves and even entire yield curves (through tools like forward guidance and yield curve control).
• Reserve Regime: Shift from scarce reserves (just enough liquidity to meet banks’ needs) to permanently ample reserves (an abundance of liquidity by design).
• Central Bank Balance Sheet: Shift from a passive, minimal balance sheet to an active, massive balance sheet that is now a structural feature of the economy.
• Market Role of Central Bank: Expansion from Lender of Last Resort (backstopping bank funding) to Dealer of Last Resort (backstopping entire markets by buying assets outright when private dealers won’t).
• Primary Policy Constraint: Shift from being limited by expected inflation and macro overheating to being limited by financial system solvency and liquidity conditions (the need to avoid crashing bond/credit markets).

Act I: The Old Regime

Before delving into the mechanical operations of QE, it is important to establish the theoretical underpinnings of the old monetary regime. A key insight is that money is not a flat, uniform stock of value; instead, the monetary system is a stratified hierarchy of liabilities. The efficacy (and limitation) of QE depends entirely on where in this hierarchy an intervention occurs.

Why Old Textbook Money Models No Longer Apply

For decades, economics textbooks taught a money multiplier model in which central banks inject base money (reserves) that banks then multiply into loans and deposits. In modern banking, this causality is reversed. Money is endogenous as commercial banks create new money in the form of deposits whenever they extend a loan. They do not lend out reserves or physical cash; instead, when a bank makes a loan, it simultaneously creates a matching deposit in the borrower’s account (new money out of thin air).

In this framework, the quantity of central bank reserves does not directly constrain bank lending. Banks lend based on profitable opportunities and then acquire whatever reserves are needed later to settle payments. In the pre-2008 scarce-reserve system, the central bank accommodated banks’ reserve needs to maintain its interest rate target. In today’s ample-reserve system, the reserve constraint is even less relevant. Reserves are interbank settlement assets; a bank can’t lend its Fed reserve balance to a household or business. Thus, expanding the monetary base (M0) via QE does not automatically expand broader money (M2) or credit in the real economy. The true constraints on lending are banks’ capital requirements, leverage limits, and the demand from solvent, creditworthy borrowers; not the volume of reserves. Understanding this loans create deposits reality is essential to explaining why the Fed’s multi-trillion-dollar QE programs after 2008 did not unleash runaway consumer inflation: those reserves stayed largely within the banking system rather than flooding into Main Street.

The Stratification of Liabilities and the Dealer Function

A more nuanced conceptual model of the monetary system is Perry Mehrling’s Hierarchy of Money. In this view, financial instruments form a pyramid ranked by their liquidity and acceptability for settling obligations, with the moneyness of liabilities decreasing as you move down the pyramid. Crucially, each layer in the hierarchy relies on convertibility into the layer above it to maintain confidence.

• At the very top are the liabilities of the state: central bank reserves and currency. This is outside money, the final settlement asset for banks. It carries no default risk in nominal terms and is the ultimate form of liquidity: the Federal Reserve ensures that one dollar of reserves or cash is always worth one dollar in payment. The central bank stands as the dealer for this highest-quality money, ready to swap reserves for other forms of money to maintain parity (e.g. supplying reserves or cash as needed to keep the system functioning at par: defined in detail in the next section).
• Just below are commercial bank deposits (the main form of inside money created by private banks). A deposit is effectively a promise by a bank to pay the bearer in official money (currency or reserves) on demand. Banks, together with deposit insurance and the central bank’s lending facilities, work to maintain a 1:1 convertibility between deposits and currency. This is not a natural law but a policy choice and an ongoing exercise in confidence: it’s the reason you believe the money in your checking account is as good as cash. The banking system performs a continuous dealer function here as banks trade reserves and collateral in money markets to ensure payments clear at par, all backstopped by the central bank’s lender-of-last-resort support when needed.
• Further down the hierarchy are various forms of credit and securities: Treasury bonds, corporate bonds, commercial paper, asset-backed securities, equities, etc. These are essentially promises to pay bank deposits (or currency) in the future, and they carry credit risk, duration risk, and liquidity risk. In good times, many of these instruments can trade nearly like money (for example, overnight repo loans treat Treasury bonds as near-cash collateral). In crises, the pyramid contracts and instruments that were treated as money-like suddenly lose that status. Investors caught holding lower-tier claims scramble to convert them into higher-tier money (selling assets for cash, drawing on credit lines, demanding currency). This flight to quality forces a rapid upshift: everyone wants the safety of bank deposits, then ultimately the safety of Fed reserves and T-bills.

When this dash for cash occurs, it’s essentially a run up the hierarchy, and only the central bank can respond at scale. To prevent a collapse of the pyramid, the Fed must expand its balance sheet by creating reserves to buy or lend against lower-tier assets and thereby providing the high-octane liquidity (top-tier money) that the system is demanding. This is how the central bank’s role evolves from merely setting interest rates to backstopping the entire hierarchy as a market-maker of last resort in crises.

The Four Prices of Money: An Integrated Constraints Framework

Another useful lens for understanding the modern central-bank constraints is the Four Prices of Money framework (articulated by Perry Mehrling and elaborated by Zoltan Pozsar). These four prices can be thought of as exchange rates between different forms of money and assets, and each represents a potential pressure point that can force central bank intervention. They are: Par, Interest, Exchange Rate, and Price Level.

1. Par – The Price of Convertibility (Trust in Banking). This is the price of private money in terms of public money, essentially the exchange rate between a dollar of bank deposit and a dollar of central bank liability (cash or reserves). In normal conditions, this price is fixed 1:1 by design; a $100 deposit at JPMorgan is valued at $100 in cash. Maintaining par convertibility is absolutely fundamental to financial stability. However, history shows that par is a policy choice, maintained by continuous support and intervention when needed. When Par breaks, it means a dollar in the banking or shadow-banking system is suddenly worth less than a government dollar, triggering panic. A classic example occurred on September 16, 2008: the $60 billion Reserve Primary money market fund broke the buck (its stable $1 shares fell to $0.97) after Lehman Brothers’ bankruptcy vaporized some of its holdings. A supposedly cash-equivalent claim was suddenly worth less than a dollar, shattering trust overnight. This breach of par triggered a run on money market funds and short-term credit markets. The Fed was forced to intervene as not just as lender of last resort but effectively as the dealer of last resort, extending guarantees and liquidity to prop up the value of these private money claims and stop the run. In a regime of financial dominance, defending Par (the banking system’s promise that deposits and money-fund balances are safe and convertible) often becomes priority number one, inflation consequences second fiddle.
2. Interest – The Price of Time (Intertemporal Exchange Rate). This is the price of money today in terms of money tomorrow: what we typically call the interest rate or the time value of money. It exists as a schedule of rates across maturities (the yield curve). In the hierarchy of money, interest is the premium for not having final settlement now: if I need money today and only have future money, I must borrow and pay interest. Central banks traditionally target the overnight risk-free interest rate (the Fed funds rate), hoping to influence the entire curve of interest rates indirectly. But the transmission is imperfect because the landscape of interest rates is a layered cake of different markets and risk premia: overnight unsecured rates, secured repo rates, term lending rates, etc., each with spreads between them. QE allows the central bank to influence interest rates more broadly by flooding the system with present money (reserves), which pushes down short-term rates and, via portfolio rebalancing, compresses longer-term yields as well. By making cash abundant and yielding next to nothing, QE tries to force investors out of the safety of cash and into longer-term or riskier assets effectively lowering the price of waiting (interest) to stimulate investment and spending.
3. Exchange Rate – The Price of Foreign Money. This is the price of one currency in terms of another (e.g. yen per dollar). It links a domestic monetary hierarchy to the rest of the world. The U.S. dollar sits atop the international hierarchy as the dominant reserve and trade currency, which means global finance runs on dollar liquidity. For many countries, their exchange rate vis-à-vis the dollar is the most sensitive constraint: a sudden drop in the currency can import inflation and disrupt local stability, or a dollar funding shortage can freeze their banking system. During crises, we often see a scramble for dollars globally as the U.S. dollar tends to appreciate sharply as everyone seeks the safest and most liquid asset. To counteract this, the Fed has established currency swap lines with foreign central banks. These swap facilities let the Fed temporarily lend dollars overseas (in exchange for equivalent foreign currency collateral) so that those central banks can on-lend dollars to their local banks and institutions. In effect, the Fed becomes the world’s dollar lender (dealer) of last resort, ensuring that a global dollar squeeze doesn’t force foreign banks to dump U.S. assets or fail to roll over dollar debts. By providing dollars through swaps, the Fed puts a ceiling on the price of dollars (or a floor under foreign exchange rates) during crises, stabilizing the international monetary system.
4. Price Level – The Price of Goods and Services (Inflation). This is the price of money in terms of the basket of real goods, essentially the inverse of the inflation rate. When the price level rises, the value of each unit of money falls. Central banks officially target this price (commonly aiming for ~2% annual inflation), but in Mehrling’s framework the general price level is often the residual that results from how the central bank manages the other three prices in combination with fiscal policy and real-economy shocks. In other words, inflation is what will be sacrificed to maintain stability in Par, Interest, or the Exchange Rate. A central bank under extreme stress will prioritize keeping the financial system solvent and the government funded even if it means letting inflation run higher than target.

All four prices are interconnected, and a central bank in a bind must often trade off one for another with impacts to inflation contingent on the overall macro-financial and fiscal conditions at play. For example, in 2008–09 the Fed prioritized preserving Par (stopping the banking panic and money-market runs) and lowering Interest rates via QE to stimulate the economy. But despite the massive liquidity injection, the Price Level stayed subdued as the new money mostly circulated within financial markets, and weak demand plus a broken credit channel kept consumer price inflation impact low. In 2020, the Fed again slashed Interest rates to zero and backstopped Par (through huge QE and emergency facilities), but this time fiscal policy simultaneously poured money directly into households’ accounts through massive direct consumer stimulus. New liquidity found a direct conduit to spending, and the Price Level subsequently jumped as inflation in 2021–2022 hit forty-year highs. By 2022, the Fed tried to clamp down on inflation (Price Level) by hiking Interest rates aggressively, but this induced losses on bond portfolios that threatened Par in the banking system (e.g. banks sitting on unrealized losses) and caused the Exchange Rate (the dollar) to soar, creating stress for dollar-dependent economies abroad. In a regime of Financial Dominance, the central bank is constantly juggling these four prices. Stability in one dimension often means instability in another, but when forced to choose, policymakers have shown they will let the inflation price be the sacrificed to preserve the financial system’s solvency and the government’s funding costs.

Act II: The Break in the Old Regime

The shift from the Old Regime to the New was not gradual or deliberate; it was forced by a violent fracture during the Global Financial Crisis. The collapse of the private dealer system in 2007–2009 redefined the central bank’s mission and shattered the illusion that markets could always self-stabilize under light-touch policy.

The Collapse of Par: September 2008

On September 16, 2008, the $60 billion Reserve Primary money market fund broke the buck as its supposedly rock-solid $1.00 shares fell to $0.97 after Lehman Brothers declared bankruptcy and its Lehman debt investments became nearly worthless. A supposedly cash-equivalent claim (a money market fund share) was suddenly worth less than a dollar in cash. This was a catastrophic breach of Par in the shadow banking system. Panic rippled through the financial world: if a prime money market fund wasn’t safe at $1, which institutions could be trusted at par? Investors fled en masse from money funds, commercial paper, and repo, the whole universe of inside money substitutes, and rushed toward the only havens left: hard cash and short-term outside money like Treasury bills and Fed reserves. In effect, the private money-market hierarchy collapsed; every actor tried to climb to the top tier of safety at the same time. Banks even stopped lending to each other out of fear. The circulatory system of global finance suffered a cardiac arrest as trust in convertibility evaporated.

How the Collapse of Repo, Haircuts, and Dealer Balance Sheets Broke the Old Regime

The crisis spread through the repo market, the vital plumbing of the financial system where banks and shadow banks borrow cash short-term by pledging securities as collateral. Before 2008, this market relied heavily on private dealer banks to intermediate trillions in daily lending. The stability of repo depended on haircuts, if you post $100 of bonds as collateral in a repo trade, you might receive $98 of cash (a 2% haircut) to protect the lender against a drop in the collateral’s value.

In 2008, as subprime mortgage assets became suspect, repo lenders didn’t all-out refuse to lend at first; instead, they demanded much larger haircuts on mortgage-linked collateral. Economists Gary Gorton and Andrew Metrick famously described this as a run on repo. Collateral that once carried a 0–2% haircut suddenly required 20%, 40%, 50% haircuts or was rejected entirely. In practical terms, this meant a firm that used to borrow $98 against a bond could now only borrow $80 or $50, if at all. Facing this gap, leveraged shadow banks had to liquidate assets to raise cash and in the process selling anything they could, which depressed prices even further. Falling prices then prompted lenders to widen haircuts even more. It was a self-reinforcing margin call death spiral.

Structural quirks in the U.S. repo system made matters worse. In the tri-party repo mechanism, two clearing banks (for example JPMorgan Chase and Bank of New York Mellon) sit between borrowers (dealers) and lenders (money market funds, etc.). Each morning, the clearing bank would unwind the previous day’s repos essentially giving lenders back their cash and returning collateral to the dealers, on the assumption that dealers would obtain new funding by the afternoon to repay the clearing bank. This process created a huge intraday credit exposure for the clearing banks; for several hours daily, the clearing bank was implicitly funding the entire dealer inventory until new repo loans settled.

When Lehman faltered, its clearing bank (JPMorgan) grew unwilling to extend this intraday credit, fearing Lehman might fail to secure new funding by day’s end. JPMorgan demanded more and more collateral from Lehman to continue the unwind. The situation was dire: if a clearing bank refused to unwind a dealer’s repos one morning, that dealer would be in instant default (unable to return cash to yesterday’s lenders). This was a single-point-of-failure risk that few appreciated beforehand. It meant the fate of major dealers hinged on the confidence and capacity of one or two private institutions, a tenuous arrangement in a panic.

As repo haircuts surged and clearing banks pulled back, private dealer balance sheets shrank at an alarming rate. Firms like Lehman, Bear Stearns, and Merrill Lynch effectively lost their lifelines. In a matter of weeks, the traditional market-making and intermediation capacity in the system collapsed, creating a vacuum that only an expansion of the public balance sheet through central bank intervention could fill.

From Lender to Market Maker of Last Resort

Confronted with this unprecedented market breakdown, the Federal Reserve had to go far beyond its classic role as Lender of Last Resort (LOLR). Under Bagehot’s 19th-century dictum, a central bank as LOLR lends freely at a penalty rate against good collateral to solvent banks essentially providing short-term loans to tide banks over any liquidity squeezes. In 2008, the Fed certainly did that (massively expanding its discount window lending and emergency facilities for banks), but that wasn’t enough. The crisis had rendered certain markets completely illiquid; there were fundamentally solvent institutions and assets that simply could not find buyers at any reasonable price. This required the Fed to become a Dealer of Last Resort (DLR), meaning it stood ready to buy assets outright to provide a price floor.

The Fed and Treasury rolled out an alphabet soup of facilities: the Commercial Paper Funding Facility (CPFF) to purchase commercial paper, the Primary Dealer Credit Facility (PDCF) to lend to securities firms, the Term Securities Lending Facility (TSLF) to swap Treasuries for illiquid collateral, and ultimately large-scale purchases of agency mortgage-backed securities and longer-term Treasuries (QE). In doing so, the Fed moved from simply easing banks’ cash crunch to directly intervening in asset markets. For the first time, monetary policy crossed into explicit credit allocation: for example, by buying mortgage-backed securities, the Fed was subsidizing housing finance over other sectors.

By assuming the role of dealer of last resort, the Fed crossed the Rubicon operationally. It was no longer just setting the price of overnight money; it was now actively setting prices across the credit spectrum by being willing to take assets onto its balance sheet en masse. This set the stage for the QE era that followed.

The Mechanics of QE: Asset Swaps and the Money Printing Myth

The idea of money printing is often used to describe QE, but in practice QE functions as a sterilized asset swap. Whether or not QE results in more money that the private sector can spend depends on who the central bank buys from. Consider two scenarios:

• Scenario A: Buying assets from a bank. Suppose the Fed buys a $100 Treasury bond from a commercial bank. The Fed pays by crediting the bank’s reserve account by $100. The bank, in turn, removes the $100 bond from its assets. The bank’s balance sheet size is unchanged: it has $100 more in reserves and $100 less in securities. Crucially, no deposit was created in the real economy; no one suddenly has more cash in their checking account. M2 (which includes deposits) is unaffected, even though the monetary base (reserves) has increased by $100. In essence, the banking system has swapped a long-term bond for a more liquid asset (reserves), upgrading the liquidity of its assets. QE in this case makes the banking system safer/more liquid, but it doesn’t directly hand anyone money to spend.
• Scenario B: Buying assets from a non-bank investor. Now suppose the Fed buys a $100 Treasury bond from a pension fund. The pension fund doesn’t have an account at the Fed, so the purchase is intermediated by a bank. The Fed credits the bank’s reserve account by $100, and the bank credits the pension fund’s checking account with $100 (in exchange for the bond, which the bank passes to the Fed). In this scenario, the bank’s balance sheet in fact expands by $100, as it has a new $100 reserve asset and a new $100 deposit liability (owed to the pension fund). Unlike Scenario A, a new deposit is created in the economy. M2 rises by $100 because the pension fund now has a larger bank account balance than before. QE in this example has created a new deposit at a bank but crucially this was still done in exchange for a bond which the pension fund could have sold before for cash regardless.

Even in Scenario B, note where that new deposit sits: in a pension fund’s portfolio. Institutions like pension funds or insurance companies aren’t going to run out and buy groceries just because their bank balance is higher. They will likely reinvest that cash into other financial assets. So the immediate impact is to drive asset prices up (the pension fund now has $100 cash it needs to allocate somewhere) rather than to drive consumer prices up. Moreover, banks on the other side of these transactions are still constrained by regulations: they can’t infinitely expand lending just because they have more reserves. Capital and liquidity rules (like the SLR and LCR discussed earlier) mean banks may treat those new deposits as hot potatoes to be offloaded (for instance, by encouraging customers to move into money market funds). In the absence of willing borrowers or fiscal stimulus, the money injected via QE ends up trapped in the financial system, circulating among banks and investors and bidding up asset prices (bonds, stocks, real estate) without fueling broad consumer inflation.

Which explains the missing inflation of the post-2008 era. Despite the Fed expanding its balance sheet by trillions, consumer price inflation stayed below target for years. QE created a lot of financial liquidity, but it mostly stayed in the financial system. By contrast, when QE was paired with massive fiscal transfers in 2020–2021 (government stimulus checks, unemployment bonuses, business grants), newly created money did enter the real economy quickly and inflation surged. Liquidity needs a conduit to generate inflation, and in 2020 the government provided one in the form of massive fiscal stimulus. It also depends on the overall macro-financial context which we explain in detail next.

Inflation: 2008 vs 2020

An ongoing misunderstanding is why the massive monetary expansion after 2008 did not spark inflation, whereas a similarly massive expansion around 2020 did. The divergent outcomes underline that context and mechanisms matter. Monetary policy does not work in a vacuum; whether it causes inflation depends on how money is injected and where it flows.

2008–2014: The Balance Sheet Recession and Broken Credit

The post-2008 period was defined by what Richard Koo terms a balance sheet recession. The Great Financial Crisis was an endogenous crisis, it originated from within the financial system, specifically a popping credit bubble in housing. When it burst, it left households, banks, and even some businesses with impaired balance sheets: assets had collapsed in value (think housing prices), but debts still remained. Suddenly, the collective priority was to pay down debt and repair balance sheets rather than to spend or invest.

This meant that even with interest rates at zero, there was little appetite to borrow. Consumers used any extra cash to deleverage (pay off mortgages, credit cards, etc.), and banks, facing both losses and new regulations, became far more conservative in making new loans. Trust in borrowers was low, and banks themselves needed to rebuild capital. The bank lending pipeline was broken or, at best, severely constricted.

Thus, when the Fed undertook QE in 2008–2014, creating ~$3.5 trillion of new reserves, most of that liquidity stayed in the financial sector. Banks ended up holding massive excess reserves, but those reserves just sat idle or were cycled through the Fed’s new interest-on-reserves framework. Broad money (M2) grew only tepidly, because banks weren’t transforming those reserves into lots of new deposits via lending. In fact, the causality reversed: instead of loans being driven by available money, loans were driven by lack of demand and banks’ risk aversion, regardless of how much money was available.

Economic growth remained sluggish, and inflation actually drifted below the Fed’s 2% target for most of the decade. The velocity of money (how frequently each dollar is spent) fell to multi-decade lows, reflecting that money was stagnating in accounts rather than circulating broadly. The Fed was essentially pushing on a string. It supplied abundant liquidity, but the private sector’s desire to borrow and spend was so weak that the liquidity mostly piled up in low-velocity venues. The experience was a powerful testament to the idea that you can’t create inflation just by creating reserves: you also need people to use that money in the real economy, and in the aftermath of a financial crisis, caution reigned.

2020–2022: The Fiscal-Monetary Fusion and the Flood

The COVID-19 shock of 2020 was an exogenous crisis:  a pandemic causing an abrupt halt to economic activity, and crucially, it hit when private sector balance sheets were generally in good shape. Households and banks weren’t overleveraged in early 2020; corporates had been borrowing a lot, but interest rates were low and debt service was manageable. The policy response to COVID recognized that a monetary response alone wouldn’t be enough to support incomes during lockdowns thus resulting in an unprecedented fusion of fiscal and monetary policy.

On the fiscal side, the U.S. government passed enormous relief packages (CARES Act, etc.) totaling around 25% of GDP within about one year. This included stimulus checks mailed directly to most Americans, enhanced unemployment benefits (in many cases more generous than wages had been), forgivable small business loans (the PPP program that essentially became grants), and aid to state/local governments. Households and firms were directly handed money on a large scale.

On the monetary side, the Fed aggressively bought Treasury bonds and MBS, in effect monetizing a large portion of the new government debt issuance. In the 12 months following March 2020, the Fed’s balance sheet grew by roughly $3 trillion, much of which was essentially financing the Treasury’s stimulus by keeping borrowing costs at zero and absorbing the flood of new bonds.

The critical difference from 2008 was that this time the newly created money went straight into the spending stream. When the Treasury sends a $2,000 check to a household, that creates a bank deposit out of thin air (a liability for a bank) and simultaneously adds reserves to the banking system (asset for a bank) when the Fed buys the corresponding Treasury that financed it. The M2 money supply (cash, checking, savings deposits, retail money funds, etc.) subsequently exploded and at one point in 2020–21 it was over 25% higher than a year before, an all-time record growth rate. And unlike in 2008, households were spending this money. In fact, personal incomes in aggregate rose during the recession due to government transfers, a far cry from the income collapse of a normal recession.

Another way to see it: the Treasury accumulated a huge balance in its account at the Fed (the TGA) during the initial borrowing phase, and then as it spent that down by issuing checks, those balances moved out into private bank accounts. This shifted liquidity from the Fed’s balance sheet to the commercial banking sector and into the hands of people and companies. By early 2022, banks were flush with deposits and eager to lend (the opposite of 2009). Consumers, flush with cash and facing reopened economies unleashed pent-up demand on everything from cars to vacations.

The result was textbook: too much money chasing too few goods, especially with COVID having disrupted supply chains. Inflation, dormant for decades, came roaring peaking over 9% CPI in the U.S. by mid-2022. In 2020, the combination of QE and direct fiscal stimulus created a potent pipeline for liquidity to flow into consumer spending. The central bank and the government were essentially acting in concert, whether officially or not: the Fed ensured cheap funding and market stability, the Treasury blasted out spending and transfers. This kind of fiscal-monetary coordination is the hallmark of an emerging regime where the two can’t be easily separated, arguably a step toward fiscal dominance, though wrapped in the guise of crisis management.

Act III: The New Monetary Regime

By the 2010s and especially after the pandemic response in 2020 a new monetary regime had fully taken hold. It is characterized by ample reserves and an implicit mandate for central banks to ensure financial market stability at all costs. Scarcity of reserves has been replaced by engineered abundance, managed through new frameworks like the floor system of monetary control. Yet this abundance coexists with regulatory constraints that keep banks cautious. In short, the modern central bank can create infinite liquidity, but must also navigate a maze of rules that make private balance sheet capacity scarce. This is Financial Dominance in operation.

The Floor System: A Corridor of Control

With trillions of dollars of excess reserves sloshing around post-2008, the Fed had to overhaul how it manages interest rates. It implemented a floor and ceiling system designed to control the price of money in an environment of superabundant reserves. The key components are:

• The Floor – ON RRP: The Overnight Reverse Repo Facility allows approved non-bank institutions (like money market funds) to lend cash to the Fed overnight at a fixed rate, receiving Treasury securities as collateral. This facility puts a hard floor under short-term interest rates because no rational lender in the market would accept a rate lower than the risk-free rate the Fed is offering. Effectively, it siphons excess cash out of private markets into the Fed’s safe harbor, ensuring that even in a glut of liquidity, overnight rates don’t fall below the Fed’s desired threshold.
• The Ceiling – SRF: The Standing Repo Facility, launched in 2021, allows primary dealers and banks to borrow cash from the Fed overnight (essentially, to repo their Treasuries with the Fed) at a rate just above the Fed’s target range. This provides an automatic safety valve against spikes in interest rates; if liquidity dries up and market rates shoot above the Fed’s comfort zone, institutions can always turn to the Fed to get cash at the SRF rate. It acts as a cap on short-term rates during times of stress (like a sudden collateral shortage or funding squeeze).
• Interest on Reserve Balances (IORB): The Fed pays interest on the reserves that banks hold at the Fed. By adjusting the IORB rate, the Fed influences banks’ willingness to lend reserves in the fed funds or repo markets. In the floor system, the IORB is set near the middle of the Fed’s target range and helps keep the effective fed funds rate within that range. Banks have little incentive to lend reserves for less than the IORB they can get safely at the Fed, which helps align market rates with policy rates.

This floor system decouples the quantity of reserves from the price of reserves (interest rate). The Fed can flood the system with reserves (for financial stability or QE purposes) without losing control of interest rates: excess cash just parks at the Fed (ON RRP or as idle reserves) rather than pushing market rates to zero. In other words, the Fed replaced the old dial-a-quantity approach with an administered rate corridor. It essentially nationalized the money markets providing a risk-free lending and borrowing outlet to retain monetary control amid balance sheet expansion. The upshot: the Fed can, in theory, expand its balance sheet as much as needed to backstop markets, while still managing to raise interest rates when it wants to tighten policy, by tweaking these administered rates.

Why Modern Bank Balance Sheets Remain Scarce Even When Reserves Are Abundant

A great irony of the New Regime is that even as the Fed drowns the system in liquidity, banks behave as if balance sheet space is more precious than ever before. This is due to post-2008 regulatory constraints that disincentivize banks from expanding their assets (even safe assets like reserves or Treasuries). Three key regulations are at work:

• Supplementary Leverage Ratio (SLR): Introduced under Basel III, the SLR requires banks to hold a minimum amount of Tier 1 capital (equity) equal to a percentage of total assets, regardless of risk. Unlike risk-weighted capital ratios, the SLR doesn’t care if an asset is a risky loan or a risk-free Fed reserve, it all counts the same. For large U.S. banks, the SLR requirement is around 5% (meaning $5 of equity for every $100 of assets). When the Fed creates reserves via QE (through scenario B above), those reserves accumulate on banks’ balance sheets and inflate the denominator of the SLR calculation. Suddenly banks need more equity capital to maintain the same ratio. But raising new equity is costly and slow. The result: banks resist holding the excess reserves. In practice, banks responded by turning away deposits. During periods of heavy QE, many banks actually encouraged corporate clients to move money into government money market funds (which then placed it into the Fed’s ON RRP facility) because those deposits were becoming a SLR burden. Excess reserves, intended to be super-liquid safe assets, became hot potatoes that banks wanted off their balance sheets. In effect, the SLR imposed a balance sheet tax on QE liquidity, meaning more reserves didn’t translate into easier lending conditions, it just meant banks bumped into a capital wall quicker.
• Liquidity Coverage Ratio (LCR): This rule requires banks to hold enough high-quality liquid assets (HQLA) to cover expected cash outflows in a 30-day stress scenario. Reserves and Treasuries count as HQLA (good for the numerator of the ratio). However, QE also creates large deposits in the system (the liability side of banks’ balance sheets), and the LCR’s assumptions treat many of these deposits as flighty. For instance, a large corporate or hedge fund deposit is considered non-operational and is assigned a high runoff rate (up to 100% of that deposit is assumed to leave in a stress scenario). So if a bank takes on a $1 billion deposit from a pension fund that just sold bonds to the Fed, the LCR might assume nearly that full $1 billion could vanish in 30 days, meaning the bank needs an extra $1 billion of HQLA on hand. In practice, that often means simply holding the proceeds as idle reserves or T-bills, effectively tying up the cash in the bank’s own liquidity buffer. This greatly diminishes the usable portion of QE liquidity. Banks see large short-term deposits not as deployable funds, but as potential liabilities that require one-for-one liquidity backing. Thus, QE funds often get siloed by LCR, quarantined to satisfy regulatory ratios rather than flowing into loans or investments.
• Net Stable Funding Ratio (NSFR): The NSFR complements the LCR by requiring banks to maintain a stable funding profile relative to their assets over a one-year horizon. It assigns required stable funding factors to assets and available stable funding factors to liabilities. A quirk of NSFR is how it treats repo and reverse repo. If a bank lends cash in a repo (i.e. takes collateral and gives cash, an asset on its books), that typically requires some stable funding (because the asset might roll over). But if the bank itself is borrowing short-term funds (like running repos on the liability side), those funds count as having zero stability if very short-term. The net effect: doing matched-maturity repo (borrowing overnight to lend overnight) is penalized. A bank might get a 0% ASF (Available Stable Funding) for the overnight borrowing but have to hold, say, 10% RSF (Required Stable Funding) for the loan. This asymmetry makes it costly in NSFR terms to intermediate repo, especially low-margin trades. Essentially, NSFR discourages banks from performing their traditional dealer role of matching short-term lenders and borrowers, because it views short-term funding as inherently unstable. Post-2015, many banks pulled back from repo market-making and let their balances shrink, contributing to episodes of illiquidity (like the repo spike in 2019). It shifted more intermediation to the Fed or to non-bank entities less constrained by NSFR (which can themselves be fragile).

In summary, while the Fed can create unlimited reserves, the banking system’s capacity to absorb and utilize those reserves is limited by these regulatory guardrails. Bank capital and stable funding have become scarce commodities. The result is a somewhat counterintuitive situation: plenty of cash in the system, but reluctance among banks to deploy it, leading to periodic cash-gluts ending up back at the Fed (via ON RRP) or sudden cash shortages when reserves get too low relative to the new normal. The central bank can lead the horse to liquidity, but it can’t make it drink (or lend) when regulatory costs loom large.

March 2020 Dash for Cash

The fragility of this new system was laid bare during the COVID-19 market panic in March 2020. Unlike 2008, it wasn’t fundamentally a credit crisis, it was a market functioning crisis, where even the safest markets seized due to a feedback loop of leverage and regulatory limits.

In early 2020, many leveraged hedge funds were engaged in the Treasury basis trade. They would buy slightly older, off-the-run Treasury bonds (which often trade at a small discount) and short Treasury futures, betting that the gap would converge. This trade was leveraged enormously with 50-to-1 leverage rations not uncommon via repo borrowing. It normally earned a low-risk retrun, given Treasuries are safe and liquid under normal conditions.

When COVID hit and a global dash for cash ensued, volatility exploded. Those hedge funds suddenly faced huge margin calls on their futures positions as Treasury prices swung wildly. To meet the cash demands, they had to fire-sell their Treasury holdings en masse. One would assume there’d be plenty of buyers for Treasuries in a crisis (the usual flight to quality), but here is where the new regime’s limits kicked in: the primary dealers could not step in to buy up all those Treasuries because of the SLR and other balance sheet constraints. Dealer balance sheets were already bloated with Treasuries from years of QE and prior trades, and taking on billions more, even temporarily, was either outside risk limits or would push capital ratios below requirements.

As a result, for a brief period the U.S. Treasury market, normally the deepest and most liquid market in the world, went no bid on certain issues. Prices dislocated: newer on-the-run Treasuries (most liquid) were trading at much higher prices (lower yields) than slightly older off-the-run Treasuries, far beyond normal spreads. Bid-ask spreads for large trades blew out multiple times larger than usual. It was an absurd situation: the world was desperate for the safety of U.S. government bonds, yet parts of the Treasury market were behaving like a stressed emerging-market bond.

The Fed’s response was swift and massive: it effectively became the buyer of last resort for Treasuries. In the span of roughly three weeks, the Fed bought over $1 trillion of Treasury securities, an absolutely unprecedented pace. This flood of Fed buying absorbed the avalanche of Treasuries that the private market could not intermediated. Liquidity was restored and yields stabilized (in fact, dropped sharply with the Fed’s intervention). The episode demonstrated a sobering reality: under the post-2008 rules, even U.S. Treasuries the bedrock collateral of the global system might not trade smoothly in a crisis without public sector intermediation. The market had become so dependent on ample balance sheet and central bank backstops that when those were lacking, it broke.

QE and the Collateral Channel

While QE is often described as easing policy by adding reserves, it has a tightening side-effect in the secured funding markets: it removes collateral that greases the system. Think of high-quality collateral (like Treasuries) as the lubricant of the financial engine, it’s used in repo, derivatives margining, securities lending, etc., often being reused or rehypothecated multiple times. The ratio of the volume of pledged collateral to the volume of underlying assets is sometimes called collateral velocity. Prior to 2008, Manmohan Singh (IMF) estimated U.S. Treasury collateral would turn over ~3.0 times (on average a single bond got reused 3 times). After the advent of QE, that number fell below 2.0.

Here’s why: when the Fed buys a Treasury bond via QE, it takes that bond out of circulation and into the Fed’s account effectively siloing it. The bond is no longer available to be pledged or lent in the market. Yes, the Fed creates an equivalent amount of bank reserves, but those reserves can’t serve the same collateral function. Reserves can’t be posted to meet a margin call; they sit on bank balance sheets, useful for payments and meeting reserve requirements but not for non-banks who need collateral. So the net effect of QE is fewer Treasuries circulating in the repo/margin markets.

With fewer Treasuries available, the system faces a safe asset scarcity problem. Banks and funds that need Treasuries to pledge will pay a premium for them or, if they can’t get them, will have to post lower-quality collateral (which might not be accepted, or will come with a bigger haircut). In periods of large QE, we’ve seen strange effects like GC (General Collateral) repo rates occasionally trading below the Fed’s RRP floor, a sign that banks were desperate to get ahold of Treasuries, even willing to lend cash at abnormally low rates, because holding Treasuries was so valuable for regulatory or safety reasons. In essence, QE injects cash but withdraws collateral, leading to potential bottlenecks in credit plumbing.

This has implications for policymakers beyond the central bank. The Treasury Department’s debt management strategy (how much to issue in short-term bills vs. long-term bonds) matters more in a QE world. Issuing more T-bills (which are high-velocity collateral, easily traded and reused) can alleviate collateral scarcity, whereas issuing mostly long-term bonds and having the Fed buy many of them exacerbates it. In recent years, we’ve seen an acute awareness of this balance, for instance, when the Fed was in QE but money market funds had no T-bills to buy, hundreds of billions rushed into the Fed’s RRP facility, basically because there was too much cash and not enough safe collateral. It’s a reminder that the composition of the balance sheet (and public debt) can matter as much as the size when it comes to market functioning (see my post on factors driving markets: https://therationalmargin.com/us-equity-outlook/).

Transmission Channels: The Mechanics of Duration Extraction

If QE doesn’t reliably work through the traditional bank lending channel (due to endogenous money and regulatory brakes), how does it then affect the broader economy? The influence of QE is felt through more indirect channels by altering financial asset prices and incentives. Three major transmission channels are often cited:

The Portfolio Balance Channel and Preferred Habitats

This is the primary mechanism central bankers emphasize. It hinges on investors not being indifferent to different asset classes, an idea formalized in the Preferred Habitat Theory (Vayanos & Vila). In simple terms, investors have favorite habitats in terms of maturity and risk. For instance, life insurers and pension funds prefer long-duration, safe assets (like long-term government bonds) to match their long-term liabilities. They don’t normally sell a 30-year bond to buy stocks just because of a small yield change; they have a strong preference to stay in their lane, unless forced out.

Enter the risk-averse arbitrageurs: hedge funds, banks, or asset managers who are willing to stray across markets to pick up bargains, but only to a limited extent. These players have capital constraints and risk limits. In a world without QE, if the government suddenly issues a lot of long-term bonds (increasing supply in that habitat), yields would rise to entice either the habitat investors to absorb more or to tempt arbitrageurs to step in and buy the excess (expecting prices to revert later). The extra yield investors demand is the term premium: compensation for the market needing to digest more interest rate risk.

When the central bank conducts QE, it acts as a gigantic, price-insensitive buyer of those long-term bonds. It removes duration supply from the market, taking bonds out of the pool that private investors need to hold. The remaining private holders are left with less interest rate risk to carry. Consequently, arbitrageurs see their inventory risk drop and require less term premium to hold bonds as yields fall relative to where they’d be without QE. This flattens the yield curve (long-term rates come down) and makes long-duration assets less attractive on a relative basis to investors. Those who get cashed out of their bonds by the Fed’s buying now have cash that they will reach for yield with, moving into slightly riskier or longer-term assets: maybe a corporate bond, maybe a mortgage-backed security, maybe stocks. That process eases financial conditions broadly as borrowing costs for businesses and households fall, and asset values rise, which can stimulate spending via wealth effects and easier financing.

In essence, QE crowd’s investors out of their safe harbors. It says to the market: “We, the central bank, are taking a bunch of these safe assets off your hands. Go do something else with your money.” That something else could be lending to the private sector, investing in new projects, or even bidding up existing assets (which can encourage new issuance of stocks or bonds). The portfolio balance channel thus relies on investor preferences and limits so it’s not arbitrage-free. When arbitrageurs have shallow pockets or strong preferences exist, the central bank’s interventions can meaningfully distort asset prices away from what they’d otherwise be, and that’s precisely the point.

Duration Extraction and the Term Premium

A closely related concept is what one might call duration extraction by the central bank. Every bond has a certain amount of duration risk: the sensitivity of its price to interest rate changes, usually measured in years. A 10-year Treasury might have about 9 years of duration, for example. Investors normally demand a term premium to hold that risk over time, especially in longer maturities where more things can go wrong (inflation, default risk in corporates, etc.).

When the Fed buys long-term bonds and holds them, it is effectively nationalizing interest rate risk. Instead of that risk being held by profit-seeking (or risk-averse) private investors who require compensation, it’s held by the central bank which is insensitive to mark-to-market losses and doesn’t require a risk premium (the Fed isn’t trying to maximize profit; it can hold bonds to maturity and has an implicit taxpayer backing). With, say, $2 trillion of long bonds moved from private hands to the Fed’s balance sheet, the private market now has a lot less aggregate duration to hold. This scarcity of duration means that investors will bid up the price of the remaining long bonds (hence yields fall) until their portfolios reach a new equilibrium.

Empirical studies of QE announcements and implementations often find that a substantial portion of QE’s impact is in lowering estimated term premiums in the bond market. In the U.S., term premium even turned negative for a while, meaning investors were accepting yields below the expected path of short-term rates, effectively paying the government for interest rate risk, which is abnormal. That was arguably a result of central bank intervention removing so much duration that the residual demand to hold safe assets far outstripped supply. Lower term premiums mean cheaper mortgages, cheaper corporate debt, and higher valuations for long-dated income streams (like stocks), all else equal.

The Signaling Channel and Shadow Rates

Beyond direct market impacts, QE also serves as a signal of policy commitment. When a central bank undertakes QE, especially if it’s open-ended or tied to conditions, it is communicating that it intends to keep monetary policy looser for longer than the market might otherwise expect. It’s a way of saying: “We will be here, holding down yields, and we’re not in a rush to tighten.” This can help shape expectations about the future path of short-term rates and reduce uncertainty. For example, the Fed’s QE3 in 2012 was explicitly open-ended (“we’ll buy $85bn/month until the labor market improves substantially”), a signal that they wouldn’t reverse course until certain outcomes were achieved.

To quantify the overall stance of policy when rates are near zero and QE is in force, economists turn to shadow rate models. The most famous is the Wu-Xia Shadow Federal Funds Rate, which essentially asks: “Given everything the Fed is doing (zero rates, forward guidance, QE), what equivalent Fed Funds rate would produce the same level of economic stimulus if it could go negative?” Because the Fed Funds rate can’t actually go negative (due to cash holding having a 0% yield floor), the shadow rate model constructs a hypothetical rate that can go negative by inferring it from bond yield movements.

Without diving into the math, the Wu-Xia model estimated, for instance, that at the height of the COVID QE blitz in 2020, the effective stance of U.S. monetary policy was like a Fed Funds rate around -5%. In 2014, after years of QE, the shadow rate was around -3% at its lowest. These numbers illustrate that QE plus forward guidance was providing a stimulus equivalent to several percentage points of negative interest rates, even though the actual policy rate was stuck at zero. In other words, QE filled the gap when conventional cuts ran out of room. It’s also a reminder that when we talk about the Fed eventually raising rates, we’re really talking about raising both the actual rate and the shadow rate, meaning stopping QE and other measures, which can be a much more significant tightening than it appears if one only looks at the Fed Funds number.

Act IV: The Trap and Global Dimensions

We have now arrived at the current state, a regime we can call the trap of Financial Dominance. The extraordinary measures taken in 2008–2009 and 2020–2021 have created a financial structure from which a smooth exit is highly improbable. It’s the policy equivalent of a roach motel: the central bank can check in (expand balance sheets, intervene deeply), but checking out (withdrawing support) is excruciating and potentially system-threatening. Every attempt at normalization has been met with market upheaval or economic stress that forces a retreat. This section explores why the trap is so hard to escape, and how it manifests not just domestically but globally.

The Mechanics of QT and the LCLOR

If QE was the entry into the hotel, Quantitative Tightening (QT) is the attempted exit. Mechanically, QT is QE in reverse: the central bank stops reinvesting bonds as they mature (and/or sells assets outright), which means the private sector must absorb more government bonds, and bank reserves get drained from the system. Yields tend to drift higher and financial conditions tighten, which is the intended outcome. However, QT is constrained by the system’s structural dependence on ample liquidity.

The key concept here is the Lowest Comfortable Level of Reserves (LCLOR): essentially the minimum level of reserves at which banks and money markets still function smoothly. Above this threshold, reserves are ample and shifts in supply have little effect on interest rates; below it, reserves become scarce and the plumbing (repo, fed funds, etc.) gets clogged as banks scramble for liquidity.

The challenge is that LCLOR is not precisely knowable in advance, and it isn’t static. It likely rises over time as the financial system and economy grows and as habits adjust to high liquidity. The Fed had a painful lesson in September 2019 on underestimating LCLOR. After years of QE, the Fed was slowly draining reserves via QT starting in 2017. By early fall 2019, bank reserves had fallen from a peak of ~$2.8 trillion to around ~$1.3 trillion. That still sounded large (far above pre-2008 levels of <$50 billion), but it turned out to at the new LCLOR. Suddenly, overnight borrowing rates spiked and the repo rate jumped to 9% in a shock move indicating a scramble for reserves and cash. Banks with excess reserves sat on them, unwilling to lend at normal rates, because their internal models and liquidity needs made them unsure about going lower. The Fed had inadvertently crossed the line where reserves were no longer ample.

Fast forward to the mid-2020s: post-COVID, the banking system had roughly $4 trillion in reserves at the peak. But the financial system’s capacity to operate with fewer reserves has likely further shrunk relative to 2019. Banks now hold even more high-quality liquid assets, market volumes are larger, and nerves are possibly more frayed. As the Fed began QT in 2022, analysts estimated the new LCLOR could be on the order of $2.5–3.5 trillion in reserves, essentially much of the QT done by 2023 might need to be permanently reversed if they hit that floor. Indeed, by late 2023, with reserves dipping near an estimated $3 trillion, the Fed signaled it would pause QT well before pre-QT projections. There was open discussion by Fed officials that they wanted to maintain a buffer of reserves above the minimum level, to avoid the ugly 2019 scenario.

In practical terms, what this means is that QT has a ceiling (or rather, a floor in reserves) and we are close to it. If the Fed tries to shrink its balance sheet beyond that point, it risks sparking another liquidity crunch that could force it to intervene (ironically) by expanding the balance sheet again via emergency operations. In other words, the central bank’s balance sheet has become structurally large and cannot shrink back to pre-crisis levels without breaking something. We’re in a world of permanently ample reserves, and the best the Fed can likely do is plateau at some higher floor. The era of scarce reserves is seemingly never coming back.

Global Dollar Funding and Financial Dominance

The trap is not merely domestic; it spans the globe due to the U.S. dollar’s outsized role. The dollar is the reserve currency and funding currency for much of the world, which means actions by the Fed can create ripple effects that trap not just the Fed, but other central banks in reactive modes.

One key manifestation is in the cross-currency swap (basis) market. Under normal conditions, covered interest rate parity (CIP) holds: borrowing dollars directly vs. borrowing in a foreign currency and swapping into dollars should cost the same. When there’s a shortage of offshore dollar funding, this parity breaks and we see a negative dollar basis, meaning it becomes more expensive to get dollars via FX swaps than it should be. This is effectively the market indicating that dollar liquidity is in high demand abroad.

We saw severe dollar funding shortages during 2008, again in late 2011 (Eurozone crisis), and again in March 2020. In each case, the Fed had to step in with or expand its swap line agreements where the Fed lends US dollars to other central banks, taking local currency as collateral, so those central banks can on-lend dollars to their banks and companies starved for dollars. During COVID, the Fed even set up temporary swap lines with smaller central banks and pumped hundreds of billions of USD liquidity abroad, which helped arrest the dollar’s spike and calmed global markets.

This arrangement has effectively made the Fed the central banker to the world. The major central banks (ECB, BOJ, BOE, SNB, etc.) now have permanent USD swap lines in place, acknowledging that dollar funding stresses are not a once-in-a-lifetime event but a recurring feature of the global system. As a result, when the Fed raises rates or implements QT, it must at least consider the global feedback loops. For instance, in 2022 the Fed’s rapid hikes contributed to the dollar’s sharp appreciation, which contributed to a UK gilts crisis (pension funds using swaps got margin calls as rates rose and the pound fell), a potential emerging-market debt crunch, and huge FX losses for countries like Japan that were forced to intervene to prop up their currency. The Fed didn’t alter its course for those reasons, but behind the scenes the swap lines were there and used modestly, and U.S. officials coordinated with other nations to mitigate volatility.

Financial dominance globally means the Fed sometimes has to act not because the U.S. needs easing, but because the world needs dollars. And if the world doesn’t get those dollars, the resulting turmoil (soaring dollar, capital outflows from EMs, crashing EM currencies, etc.) can circle back to hurt U.S. financial stability (e.g. if an EM crisis causes global banks to wobble or triggers a credit event). Thus, the Fed is in a de facto arrangement of providing a dollar safety net internationally, effectively exporting its lender-of-last-resort function. One could argue this is also part of an underlying policy of turning unlimited dollar liquidity into a tool of global financial governance, beyond domestic mandates.

The Japanese Trap: Solvency Margins and Duration Gaps

If one wants to see the endgame of these dynamics, Japan offers a cautionary tale. The Bank of Japan (BOJ) has been engaging in quantitative easing and yield curve control for decades, to the point that it now owns over 50% of all Japanese Government Bonds (JGBs) outstanding. This heavy hand has kept Japanese interest rates near zero for a very long time. But it has also induced major distortions in their financial system.

Japanese banks and, importantly, life insurance companies hold the rest of the JGBs. Life insurers in Japan have massive long-term liabilities (payouts to policyholders, often with guaranteed returns). They invested heavily in JGBs and foreign bonds to try to earn yields to meet those guarantees. Over time, they developed a significant negative duration gap: their liabilities (the insurance policies) are longer-term than their assets, meaning if interest rates rise, their assets (bonds) drop in value more than their liabilities as the liabilities’ valuations were effectively locked in at higher discount rates from the past.

Normally, you’d think higher rates help insurers (they can reinvest at better yields). But here in lies the twist: Japan’s regulatory framework, via the Solvency Margin Ratio (SMR), historically allowed insurers to value liabilities using a fixed discount rate (often assuming higher rates than reality) and required them to mark assets to market. So if the BOJ suddenly allowed JGB yields to jump, insurers’ bond portfolios would take a hit, but their reported liabilities would not shrink equivalently (since those were pegged to an outdated assumed interest rate). The result would be a plunge in their solvency ratios where on paper, some could even become insolvent, despite the long-run business maybe actually improving with higher rates.

The BOJ knows this. In effect, by pinning rates at ultra-low levels for so long, it induced a kind of hyper-sensitivity in the system to rising rates. Now the BOJ is stuck: any attempt to normalize (let yields rise) has to be gradual and carefully coordinated with regulatory changes. (Japan is moving to a new economic value-based solvency regime by 2025, which should theoretically mark liabilities to market and be more rational, but getting there without triggering a crisis is the challenge).

This is a vivid example of financial dominance at an extreme: the central bank of Japan is essentially a hostage to the solvency concerns of the institutions it was regulating. It was not just abstract inflation or employment on the scales, but the very real prospect of blowing up the life insurance sector or causing a government funding squeeze (the government benefits from ultra-low rates to finance its huge debt). So the BOJ maintained yield curve control, even as inflation ticked up, far longer than any other central bank, because it simply had no clean exit. The state (MOF and BOJ) and the financial system have merged into one giant, delicate balance sheet that can’t be easily separated.

The Swiss Trap: Balance Sheet Size and Distortions

Switzerland’s case is quite different in the cause, but similar in the outcome: a central bank with a bloated balance sheet that creates political and financial entanglements. The Swiss National Bank (SNB) spent much of the 2010s intervening to prevent the Swiss franc from appreciating too much (Switzerland is a small open economy with a perpetual strong-currency tendency during global stress). To do this, the SNB printed francs and bought foreign assets, a lot of foreign assets. The SNB’s balance sheet swelled to over 100% of Swiss GDP, an unheard-of level for a central bank outside of wartime or hyperinflation.

The SNB became, in effect, one of the world’s largest investment funds. It accumulated hundreds of billions in foreign bonds and also equities as the SNB directly owns sizable stakes in big-name companies like Apple, Microsoft, Amazon, etc. This equity buying was simply a byproduct of needing to diversify all the foreign currency they bought (you can’t put hundreds of billions into negative-yielding Euro government bonds without suffering, so they moved into stocks for some return).

For a while, this looked very smart as the SNB made profits as global markets rose. Those profits were partly distributed to the Swiss cantons (states) and the federal government, which had come to expect this central bank dividend as part of their budgets. But then 2022 happened and global equities and bonds fell sharply, and the SNB, with its huge exposure, incurred an annual loss of around CHF 132 billion. A staggering 18% of Swiss GDP. This wiped out its equity capital and forced it to suspend the regular payments to the cantons and government.

Suddenly, local politicians were up in arms as budgets had holes because the SNB lost money that would have otherwise gone to public coffers. Some demanded the SNB stop equity purchases or even unwind positions to realize gains earlier (when it had them). Others questioned the strategy altogether. The SNB, which is operationally independent, found itself at the center of political debate, not because it failed on inflation (inflation was actually quite low in Switzerland) but because its balance sheet size and risk-taking had major fiscal implications.

This highlights another facet of the modern regime: central banks venturing into quasi-fiscal roles. When a central bank massively intervenes, it can accumulate risk and even suffer losses like any bank or fund. Normally, central bank losses aren’t a big deal (they can operate with negative equity in accounting terms, as the SNB now is, and still function fine). But when those losses affect public finances (cantons didn’t get expected money) or public sentiment (“our central bank is gambling in U.S. tech stocks?!”), it challenges the notion of central bank operations being purely about monetary policy. It becomes political. The SNB example shows that even doing the right thing for your mandate (preventing currency overshoot, fighting deflation via negative rates and interventions) can store up political risk that later constrains your freedom or invites calls for oversight. The SNB now has to tread carefully as it has it started modestly shrinking its balance sheet, but any major changes will be scrutinized by a public newly awakened to the fact that the central bank is intertwined with their fiscal reality.

2024 Treasury Buybacks

A final recent example of the new paradigm can be found in a seemingly mundane development in U.S. government debt management. In 2024, the U.S. Treasury announced it would begin buybacks of older, off-the-run Treasury securities, something it hasn’t done regularly in decades (since the early 2000s). On the surface, the stated reason was cash management and debt management efficiency. But functionally, this program is aimed at bolstering Treasury market liquidity.

Over time, as Treasuries age, they tend to trade less frequently and can become less liquid, especially if many are held by long-term investors. These off-the-run bonds often trade at a slight price discount (higher yield) compared to the newest issues (on-the-run) which are in demand. Normally, primary dealers would arbitrage some of this or at least provide a market, but as discussed, their capacity isn’t what it used to be. Market participants have complained about poor liquidity in certain Treasury maturities, which is a systemic risk since Treasuries are used as collateral everywhere.

The Treasury’s buyback plan is to purchase some of these off-the-run Treasuries (injecting cash to those holders) and simultaneously issue more of the current on-the-run issues (raising the cash to do so). In essence, the Treasury is swapping illiquid bonds for liquid bonds. This helps investors and dealers by providing an outlet for harder-to-sell bonds and increasing the supply of the most liquid benchmarks.

Why is this remarkable? Because it shows the government itself stepping in to support market liquidity, a role historically played by investment banks and dealers. It’s an admission that the dealer system, hampered by capital rules and risk aversion, is no longer smoothly performing one of its crucial functions: recycling older debt and market-making. The state, which already relies on the Fed for market liquidity, is now directly engaging in market-making activities for its own debt.

In a way, the Treasury buyback scheme, the Fed’s SRF facility, and even discussions of the Fed potentially backstopping mortgage markets in stress, all these point to the same conclusion: the line between public and private balance sheets has blurred. The state apparatus (Treasury+Fed) is increasingly the only balance sheet big enough and willing enough to ensure continuous market functioning. That is the heart of the financial dominance trap: once the state starts doing this routinely, the market comes to expect it, and private actors plan around it. The state becomes the market.

Conclusion: The Crossed Rubicon

The journey from the old regime of monetary scarcity to today’s trap of financial dominance leads to an inescapable conclusion: this is likely not a temporary crisis-era mix of policies, it is more than likely the permanent architecture of the new modern financial system. The policy that began as emergency stimulus is now embedded as a structural feature. Consider the path we’ve traveled:

• Money-Printing vs Inflation: QE shattered the old notion that expanding the money supply automatically causes proportional inflation. Trillions in reserves were created, yet for a long time inflation stayed low. We learned that the context (broken banks vs. fiscal fusion) determines inflation, not just the quantity of money. The price of money has been decoupled from its quantity, something that defied orthodox expectations.
• Regulation-Induced Scarcity: Post-2008 regulatory reforms (Basel III rules like SLR, LCR, NSFR) made bank balance sheet capacity scarce and expensive. Banks could no longer leverage or make markets freely. This meant that even though central banks flooded the system with liquidity, that liquidity often got stuck. It also meant private sector intermediation in crucial markets became constrained, a fragility that would require the public sector to step in later.
• Collapse of Private Market-Making: The high cost of capital and leverage caused a collapse in private dealer capacity. Banks and broker-dealers dramatically pulled back from providing liquidity in repo, bonds, and derivatives. We saw the consequences: repo market spikes, Treasury flash crashes, basis trade unwinds. The system became prone to sudden seizures because the traditional market makers were no longer there in size.
• The Central Bank as Dealer of Last Resort: Into this gap stepped the central banks, willingly or not. The Federal Reserve and peers became the dealers of last resort, meaning they stood ready to buy assets and provide liquidity when no one else would. The Fed’s balance sheet went from a passive footnote to a primary driver of markets. Facilities like the Standing Repo Facility (SRF) are now permanent, and massive interventions are expected whenever markets wobble. The state effectively backstopped the entire financial market, not just the banking sector.
• Market Dependency on Public Support: Markets have since become addicted to the public backstop. Each attempt to withdraw support, whether through the Fed’s QT in 2018–2019, or rapid hikes in 2022, or the Bank of England’s attempt to shrink its balance sheet, has been met with turmoil that forced a policy pivot or bailout (e.g., the Fed’s repo operations in 2019, or the BOE’s emergency gilt buying in 2022 to save pension funds). Exiting these interventions without incident has proven nearly impossible. We are effectively stuck in a perpetual quasi-QE environment, even if active QE is paused, as markets expect central banks to ride to the rescue if conditions tighten too much.
• Financial Dominance Entrenched (The Fed Put): In theory, central banks could ignore financial stability and press on with fighting inflation no matter the pain. In practice, they have repeatedly blinked when faced with crises that threaten system solvency or critical market functioning. This signals a regime of Financial Dominance: where financial system considerations dominate the policy landscape, often overshadowing the inflation targeting goal. Central banks still talk tough on inflation and may act against it, but there is a limit defined by when vital financial gears start to grind or when government financing strain becomes too severe.

In this brave new world, the state is the market. The once-clear line between public policy and private finance has dissolved. Through its commitment to maintaining a floor on reserves (honoring the LCLOR constraint), deploying standing facilities like the SRF, and even direct Treasury market actions like buybacks, the government and central bank now explicitly guarantee liquidity for the debt market and by extension much of the financial system. The central bank is the largest holder of government debt; its balance sheet health is intertwined with the fiscal authority. Indeed, central bank losses (like the Fed’s operating loss from paying more interest on reserves than it earns on bonds, or the BoE’s QE losses indemnified by the Treasury) have fiscal implications. Monetary and fiscal policies are thus bound together in a feedback loop.

Looking ahead, this means that the traditional notion of an independent central bank solely focused on inflation is, for all practical purposes, obsolete. Inflation vs. Stability Trade-off: When push comes to shove, if containing inflation threatens to blow up the financial system or bankrupt the government, central banks will flinch. They will tolerate higher inflation or resort to price-suppressing measures (financial repression) rather than allow a 1930s-style collapse. High debt loads all but ensure this. In blunt terms, we are closer to a fiscal dominance environment than any time in recent memory; one where the central bank ultimately ensures the government’s solvency (through low rates or outright purchases) because failing to do so would be catastrophic.

For investors and citizens, the price signals we get from markets are no longer purely free-market outcomes. Interest rates, credit spreads, asset valuations all are heavily influenced by the expectation of if/when the central bank will step in or step back. Price is increasingly a policy tool or a policy byproduct. We saw this vividly when the Bank of England’s attempt at hands-off discipline led to a spike in yields that, via pension fund leverage, almost caused a meltdown and promptly forcing the BOE to buy bonds and drop yields regardless of inflation readings. That’s a microcosm of the future: policy will let markets go only so far before intervening, and markets know this, and will behave accordingly.

The central alchemy of turning debt into apparent liquidity and risk-free assets via central bank magic has a dark side, but it’s now the permanent state of affairs. Escaping this architecture would require painful structural reforms, likely significant austerity or defaults to reduce debt, and a willingness to withstand severe short-term crises to reset the system. Those options are not palatable in any modern democracy except in extremis. Thus, we muddle forward in the new regime.

To conclude, investors and policymakers alike must internalize that we are for now permanent residents of this new world. It’s a world where central banks have gone from last-resort lenders to all-the-time market makers, where government debt is effectively being partially monetized in stealth, and where inflation, growth, and yields will be managed within political constraints of debt sustainability. It’s not necessarily dystopian and it can function, as the past decade shows, but it is fundamentally different from the pre-2008 paradigm. In this world, market outcomes are highly conditioned on policy, and navigating it requires understanding that the traditional signals (like a rising yield meaning tighter policy ahead) might be short-circuited by intervention.

In a market where the regulator is the largest participant, the price signal is no longer a pure discovery of value, it is, in large part, a reflection of policy choices. The architecture of the new regime is here to stay, and we must all adapt to its reality.

Implications for Portfolios

The regime of Financial Dominance necessitates a rewriting of the institutional investment playbook. Many traditional assumptions (like always using Treasuries for crisis hedging, or expecting central banks to prioritize inflation above all) may no longer hold. Investors need to navigate an environment where market prices are heavily influenced by policy decisions and where the risk/return characteristics of assets can shift in unexpected ways due to that policy interference.

Duration, Convexity, and the Volatility Feedback Loop

In fixed income, one hidden source of instability today is negative convexity especially stemming from the mortgage-backed securities (MBS) market. Unlike a plain vanilla Treasury bond, an MBS gives the borrower (homeowner) an option to prepay the loan. This means when interest rates fall, homeowners tend to refinance and pay off their old loans faster, causing MBS investors to get their principal back sooner (i.e., the bond’s duration shortens). Conversely, when interest rates rise, refinancings dry up and those mortgages stay outstanding longer than expected (the bond’s duration extends). In short, MBS have adverse convexity: their duration increases in a rising-rate environment and decreases in a falling-rate environment.

This wouldn’t be a big problem if investors just held MBS to earn the spread, but major MBS holders (like mortgage REITs, hedge funds, and banks) often hedge their interest rate risk. When rates start rising and their MBS duration extends, they suddenly find themselves under-hedged (their assets are more sensitive to rates than their hedges). To compensate, they sell Treasuries or pay fixed in swaps to increase their hedges, which adds upward pressure on long-term rates. But that selling of Treasuries causes rates to rise further, which again extends MBS durations, forcing more hedging sales. This can become a self-reinforcing spiral: a convexity hedging loop that amplifies bond market volatility. Historically, episodes like 1994’s bond massacre and the 2003 spike in rates were exacerbated by such dynamics in the mortgage market.

In the current regime, the Fed (which after the GFC became a huge MBS holder) is pulling back by letting its MBS holdings run off. That means more of this convexity risk is in private hands now. The absence of the Fed as a steady volatility dampener in the mortgage market implies that interest rate volatility can be higher, and rate sell-offs can become more violent than they used to. Investors should be prepared for larger swings in yields and consider the impact of convexity. This could mean demanding higher risk premia for holding MBS, being quicker to deleverage when volatility starts rising, or using options to protect against tail scenarios. In practical portfolio terms, one might avoid over-concentration in negative convexity assets or at least size them with the understanding that they can behave in non-linear ways in stressed markets.

Stock-Bond Correlation and Portfolio Construction

Perhaps the most critical shift for strategic asset allocation is the breakdown of the negative stock-bond correlation that defined the last two decades. From roughly 2000 to 2020, U.S. stocks and Treasuries were negatively correlated: when stocks fell in a risk-off move, Treasuries would typically rally (yields fall), providing a natural hedge. This underpinning made the 60/40 stock/bond portfolio a robust, low-volatility strategy. It hinged on inflation being low and stable so that bad economic news brought expectations of lower inflation or easier Fed policy, which was good for bonds.

In the new environment of higher inflation and massive public debt, that correlation has flipped positive at times. We saw a painful illustration in 2022 as inflation spiked to multi-decade highs, and the Fed embarked on rapid rate hikes. The result was that and both stocks and bonds sold off in tandem (bonds had one of their worst years in history, right when stocks were also deep in the red). Why? Because the primary risk factor became inflation and the Fed’s response to it. In such regimes, good economic news (strong growth) can be bad for both stocks and bonds, it implies more inflationary pressure or more Fed tightening, hurting bonds directly and stocks via the prospect of economic overheating or a heavier policy hand. Conversely, bad economic news (recessionary signals) might eventually be bullish for bonds (since the Fed would cut rates), but initially it also slams stocks and can be chaotic for credit and if there are concerns about financial stability, even safe bonds might cheapen until the Fed actually pivots.

The implication is that traditional diversification needs rethinking. If your Treasury holdings can drop at the same time as your equities, you don’t have much of a hedge. Investors, especially those managing multi-asset portfolios or using risk-parity type approaches, will need to incorporate alternative diversifiers and hedging strategies. Some approaches include:

• Active Duration Management: Instead of always holding long-duration bonds as a hedge, investors may need to dynamically adjust duration. In periods where inflation risks dominate and central banks are hiking, holding less duration (or even being short duration) can protect the portfolio. Strategies like yield-curve steepeners (long short-term Treasuries, short long-term Treasuries) can profit if long-term yields rise faster due to term premium expansion, which is plausible when debt levels are high and central banks are not outright buying long bonds.
• Real Assets and Inflation Hedges: Adding assets that are positively correlated with inflation can fill the hedge role that bonds used to play. Commodities are a prime example, they tend to do well in late-cycle or inflationary shocks (as seen in 2021–2022 when commodity indices surged). Gold is another classic anti-paper asset; it may not have a consistent correlation with stocks or bonds, but in scenarios of fiscal/financial repression or currency debasement (real yields deeply negative), gold can shine as a store of value. These assets don’t produce yield, but that’s the trade-off for their diversifying properties. They can help offset a simultaneous drawdown in stocks and bonds driven by inflation or currency worries.
• Volatility and Tail Risk Strategies: With bonds less reliable as crisis balancers, long volatility strategies become more important. This could mean maintaining allocations to strategies like long equity index put options, long VIX calls, or long Treasury volatility (options on Treasury futures), which pay off during market stress when volatility jumps. It could also involve systematic tail-risk funds or strategies that use trend-following (Managed Futures CTAs, for example, often profit in sustained market sell-offs by shorting equities or bonds). In 2022, some trend-following funds had one of their best years, precisely because they could short bonds and capture that tail risk profit that a 60/40 investor lost. Incorporating these kinds of tail hedges or diversifiers can cushion portfolio losses when both equities and standard fixed income are struggling.
• Re-evaluating Equity and Credit Exposures: In a financially repressed environment, one might favor equities with pricing power, real asset exposure, or inflation linkages (e.g., some commodity producers, infrastructure, REITs with short lease durations). Credit is tricky as inflation erodes real value for creditors, and yields might lag inflation, meaning negative real returns. However, if central banks cap yields to save governments (financial repression), some credit, like shorter-term high yield or emerging market debt, might benefit from carry if defaults remain in check. It becomes a more nuanced game of selecting where policy support might indirectly prop up certain assets. Equity duration (sensitivity to rates) also matters: high-growth, long-duration equities (tech) got hit hard when rates rose; more value or dividend-oriented stocks fared better. A balanced equity portfolio might lean more on value/cash-flow generative companies if we expect persistently higher discount rates.

Overall, investor strategy must become more tactical and multi-faceted. The days of set-and-forget 60/40 allocation counting on central bank-independent cycles may be over for a while. Now one must pay attention to policy regimes: Are we in an easing mode where the Fed will backstop everything (favors long risk assets, long duration)? Or in a tightening/stagflationary mode where the Fed’s hands are tied (favors cash, commodities, and short duration)? Or even in a yield-curve-control environment (favor equities and perhaps long bonds since central bank suppresses yields, but watch currency debasement, favor gold)?

In sum, under financial dominance, flexibility and diversification beyond the traditional are key. Portfolios likely need a spread of assets including commodities, maybe some inflation-indexed bonds, gold, and dedicated tail hedges, alongside the usual stocks and bonds. Risk management becomes more central: volatility targeting, drawdown control. Because the old reliable correlations may not save you. And perhaps most importantly, keeping an eye on the policy narrative is as crucial as tracking earnings or valuations. In markets that are policy-driven, understanding the reaction function of central banks and governments (and their constraints) becomes part of the investor’s job.

Sources

1. Bagehot, Walter. Lombard Street: A Description of the Money Market. London: Henry S. King & Co., 1873. Public domain. (Available via Project Gutenberg).
2. Buiter, Willem H., and Anne Sibert. “The Central Bank as the Market-Maker of Last Resort: From Lender of Last Resort to Market-Maker of Last Resort.” VoxEU (CEPR), August 13, 2007.
3. Federal Reserve Bank of Chicago. Modern Money Mechanics: A Workbook on Bank Reserves and Deposit Expansion. Chicago: FRB Chicago Public Information Center, originally 1961 (rev. ed. 1992).
4. Gorton, Gary B., and Andrew Metrick. “Securitized Banking and the Run on Repo.” Journal of Financial Economics 104, no. 3 (2012): 425–451. (NBER Working Paper No. 15223, August 2009).
5. Koo, Richard C. Balance Sheet Recession: Japan’s Struggle with Uncharted Economics and Its Global Implications. Singapore: John Wiley & Sons (Asia), 2003.
6. McLeay, Michael, Amar Radia, and Ryland Thomas. “Money Creation in the Modern Economy.” Bank of England Quarterly Bulletin 54, no. 1 (2014): 14–27.
7. Mehrling, Perry. The New Lombard Street: How the Fed Became the Dealer of Last Resort. Princeton, NJ: Princeton University Press, 2011.
8. Mehrling, Perry. “The Inherent Hierarchy of Money.” In Social Fairness and Economics: Economic Essays in the Spirit of Duncan Foley, edited by Lance Taylor, Armon Rezai, and Thomas Michl, 394–404. London: Routledge, 2013. (Pre-print available from Boston University archive.)
9. Phillips, A. W. “The Relation Between Unemployment and the Rate of Change of Money Wage Rates in the United Kingdom, 1861–1957.” Economica 25, no. 100 (November 1958): 283–299.
10. Pozsar, Zoltan, Tobias Adrian, Adam B. Ashcraft, and Hayley Boesky. “Shadow Banking.” Federal Reserve Bank of New York Staff Report No. 458, July 2010. (Published in FRBNY Economic Policy Review 19, no. 2 (2013): 1–16.)
11. Pozsar, Zoltan. “Money, Commodities, and Bretton Woods III.” Credit Suisse Global Money Notes (Investment Strategy), March 31, 2022. (Public repost PDF available online.)
12. Sargent, Thomas J., and Neil Wallace. “Some Unpleasant Monetarist Arithmetic.” Federal Reserve Bank of Minneapolis Quarterly Review 5, no. 3 (Fall 1981): 1–17.
13. Taylor, John B. “Discretion versus Policy Rules in Practice.” Carnegie-Rochester Conference Series on Public Policy 39 (1993): 195–214.
14. Wu, Jing Cynthia, and Fan Dora Xia. “Measuring the Macroeconomic Impact of Monetary Policy at the Zero Lower Bound.” Journal of Money, Credit and Banking 48, nos. 2–3 (March–April 2016): 253–291. (Earlier version presented at IMF Research Conference, 2015).