From Zero Interest Rates to QE

The first five lessons built the architecture: the economy as a network of balance sheets (Lesson 1), money (Lesson 2), credit (Lesson 3), banks (Lesson 4) and sectoral balance sheets (Lesson 5). Lesson 6 then asked what a central bank actually controls, and separated direct control from influence. This lesson takes the next step: what a central bank does when the instrument it controls most directly, the policy rate, has run out of room.

1. Lessons 1–5: balance sheets → money → credit → banks → sectoral balance sheets
   ↓
2. Lesson 6: central banks and what they actually control
   ↓
3. Lesson 7: from the policy rate to central-bank balance-sheet policy and QE

The deeper lesson is not simply that central banks bought bonds. It is that central banks began actively changing the structure of private-sector balance sheets and financial-market portfolios in order to influence financial conditions. That distinction is essential for understanding modern asset markets, and it is the bridge to the asset-price lessons that follow.

For decades the conventional framework was straightforward. A central bank adjusted its policy interest rate: higher rates generally tightened financial conditions and lower rates generally eased them. The policy rate was the principal instrument. The mechanism is intuitive, because lower rates can reduce borrowing costs, mortgage rates, corporate financing costs, discount rates, government bond yields and the required returns on financial assets, and can encourage borrowing, investment, consumption and portfolio reallocation.

The global financial crisis exposed a limit. Central banks cut rates aggressively, and in many advanced economies policy rates approached zero. Yet activity stayed weak, inflation stayed subdued, credit conditions remained fragile and markets continued to experience severe stress. The constraint is practical: nominal interest rates cannot normally be lowered indefinitely, because at sufficiently low rates cash becomes an increasingly attractive alternative to negative-yielding instruments. This is the effective lower bound. Its precise level depends on the monetary system and operating framework, and it was approached in different forms in the United States, Japan and Europe.

The problem is simple to state. The central bank may want to provide additional accommodation while the conventional instrument has very little room left to fall. The answer was to rely less exclusively on the short-term rate and more on the balance sheet.

The two instruments are related, but they are not identical. Traditional monetary policy is a price-of-money policy; QE adds a balance-sheet dimension.

Quantitative easing is a monetary-policy strategy in which a central bank purchases financial assets on a large scale to influence financial conditions beyond the conventional short-term policy rate. Depending on the institutional framework, the assets can include government bonds, agency securities, mortgage-backed securities and other eligible securities. Design differs across jurisdictions, but the balance-sheet mechanism is common.

A simplified central-bank balance sheet, introduced in Lesson 6, has securities, foreign-exchange reserves and loans on the asset side, and currency and commercial-bank reserves on the liability side. When the central bank buys securities, its balance sheet expands: it acquires an asset, and the private financial system receives a corresponding increase in reserves or another settlement liability. The private sector exchanges one type of asset for another.

Illustration (hypothetical). Suppose a central bank purchases €100 million of government bonds from a commercial bank. Only the changes are shown.

The commercial bank has exchanged bonds for reserves; the central bank has acquired the bonds. The government has not received €100 million of new spending power. What changed is the composition of financial assets within the private sector, and that composition is where the policy acts.

“The central bank prints money” is a useful shorthand for a broad audience, but it misleads. QE does create additional central-bank money in the form of reserves. Reserves, however, are not household deposits. They are liabilities of the central bank and assets of commercial banks, and they exist within the banking and settlement system. A bank that receives additional reserves does not automatically pass them to households.

These are different layers of the monetary hierarchy (Lessons 2 and 4). When the seller is a non-bank investor, such as a pension fund, the seller’s bank deposit also rises as the bank credits the account and receives reserves, as Lesson 6 noted. Even then the central bank does not credit households at large. QE therefore does not mean: central bank creates reserves, households receive money, households spend it. The actual transmission operates through bond yields, portfolio allocation, term premia, expectations, liquidity, asset valuations, credit conditions, exchange rates, wealth effects and financial-sector balance sheets.

The central mechanism of QE is portfolio rebalancing. Investors hold diversified portfolios of government bonds, corporate bonds, equities, cash, real estate and other assets. When the central bank buys a large quantity of government bonds, the private sector holds fewer of them, but investors still hold the same financial wealth and must decide how to allocate it. Some will buy longer-duration bonds, corporate bonds, equities, foreign securities or credit instruments.

Duration measures, in simplified terms, the sensitivity of a bond’s price to changes in interest rates. A long-term government bond carries a significant amount of interest-rate risk. When a central bank buys large quantities of long-duration securities it removes part of that duration from the private market and absorbs it onto its own balance sheet. This is duration compression.

It matters because of the term premium. A long-term yield can be thought of, conceptually, as the expected path of future short-term rates plus a term premium: the extra compensation investors require for holding a longer-duration bond rather than repeatedly holding short-term instruments. The term premium reflects interest-rate and inflation uncertainty, duration risk, liquidity, investor demand, risk aversion and the supply of long-term securities. If the central bank removes a substantial quantity of duration, investors may require less compensation for holding what remains, and the term premium can decline. That is one reason QE can lower long-term yields when the policy rate is already close to zero.

The policy rate acts mainly at the short end of the curve. QE can act on the long end: five-, ten- and thirty-year yields, mortgage rates, corporate borrowing costs and the discount rates used in valuation. The central bank is attempting to influence the term structure of interest rates, not only the overnight rate.

Under severe stress, liquidity can deteriorate rapidly: investors become reluctant to transact, bid-ask spreads widen, funding markets become unstable and participants try to sell at once. The feedback loop runs from liquidity deterioration to forced selling, falling prices, weaker collateral, higher margin requirements and further selling. Large-scale purchases provide a major source of demand that can interrupt it.

QE should not be confused with every form of central-bank liquidity provision. A central bank can also lend through standing and emergency facilities, repo operations and collateralised lending (Lesson 6). QE is specifically the large-scale purchase of assets as a monetary-policy instrument.

A related effect is scarcity. If investors strongly prefer safe long-duration government bonds and the central bank removes a substantial quantity from the market, the remaining bonds become relatively scarce, with higher prices and lower yields. This scarcity channel is closely related to portfolio rebalancing and duration compression: the central bank is changing not only the quantity of reserves but the composition of assets available to private investors.

QE does not need to involve direct purchases of corporate bonds or equities to influence their prices. If government yields fall significantly, investors searching for return shift toward investment-grade and high-yield credit, equities, emerging-market assets and other riskier securities. Corporate spreads can narrow, borrowing costs can decline, and firms may find it easier or cheaper to refinance or issue debt. The central bank does not need to control every asset directly: it changes their relative scarcity and attractiveness.

The link to valuation is the discount rate. In a deliberately simplified relationship, asset value is approximately future cash flows divided by a discount rate. If the discount rate falls, the present value of future cash flows tends to rise, most strongly for long-duration assets: long-term bonds, growth equities, infrastructure, real estate and long-dated corporate projects. QE can lower the risk-free component and, through portfolio rebalancing, compress risk premia, producing a broad repricing. This does not mean QE mechanically determines equity prices: they still depend on earnings, growth, productivity, margins, taxation, regulation and expectations. But the discount-rate component of valuation can be shifted materially. Lesson 8 develops this channel in detail.

QE also has a signalling dimension. A large asset-purchase announcement can be read as evidence that policy will stay accommodative, that short-term rates may remain low, that the central bank is willing to support financial conditions, that it expects prolonged weakness, or that it will respond aggressively to stress. Expectations themselves move asset prices, so the announcement can have an effect before purchases are complete.

QE also changes the perceived reaction function. Markets begin to ask under what conditions purchases will increase, when they will stop, whether maturing securities will be reinvested, whether the balance sheet will be allowed to shrink, how QE interacts with the policy rate, whether the central bank will intervene in stress, and what level of inflation would prompt it to withdraw accommodation. The central bank therefore influences markets not only through transactions but through the expected future path of its balance sheet and policy rate.

When a central bank buys a government bond from a commercial bank, the bank’s asset changes from a government bond to a central-bank reserve. It has not necessarily become richer in a net-worth sense: the composition, liquidity and risk characteristics of its assets have changed, without mechanically creating an equivalent amount of bank equity.

More reserves therefore do not automatically mean more bank lending. Banks do not generally wait for QE-created reserves before deciding whether to lend. Lending is constrained mainly by the economics and risk of lending, as Lesson 4 showed. A bank asks:

• Is the borrower creditworthy?
• Is the loan profitable after funding costs?
• Does the bank have sufficient capital?
• What is the expected loss, and what collateral is available?
• What is the regulatory capital requirement?
• What is the expected demand for credit?
• What are funding and liquidity conditions?

QE can influence some of these conditions, but it does not remove the constraints. It is not a mechanism for multiplying reserves into loans; it changes the broader financial environment in which banks, investors and borrowers operate.

Rising financial-asset prices can affect the real economy through wealth effects. Households that own equities, pension assets, bonds or real estate may see measured wealth rise, which can influence consumption, confidence, investment and borrowing. The mechanism is not universal, because the distribution of asset ownership matters. A rise in equity prices has a different effect when financial assets are concentrated among high-wealth households than when ownership is broad, and households that depend mainly on wages or cash savings experience different effects.

Asset-price effects are therefore also distributional effects. A more useful question than whether aggregate wealth increased is whose balance sheet improved, whose financing costs declined, and who owned the assets whose prices rose. Aggregate statistics can conceal these balance-sheet differences, which is a limitation of simplified monetary-policy models.

Financial markets are globally connected. If a major central bank lowers yields and compresses term premia, global investors may search for higher returns elsewhere: foreign bonds, equities, emerging markets, higher-yielding currencies and corporate credit. The result can be a global easing of financial conditions. The reverse can also occur. When major central banks tighten, global yields rise, the dollar may strengthen, capital flows can reverse, emerging-market financing conditions can deteriorate and foreign-currency borrowers become more vulnerable.

QE is therefore not merely a domestic instrument. It can become part of the global financial cycle, a theme developed later in this manual.

Large-scale government-bond purchases create an important link between monetary and fiscal policy. The central bank becomes a significant holder of government liabilities. The government continues to owe the debt, but the composition of the private sector’s claims changes. QE does not mean that government debt has disappeared: the liability remains within the consolidated public sector, although its ownership and maturity structure have changed.

This matters for government financing, debt maturity, interest expenditure, central-bank remittances and monetary–fiscal interactions. The two authorities remain institutionally distinct, but their balance sheets become increasingly interconnected.

QE does not eliminate risk. When the central bank buys duration from the private sector, private portfolios carry less duration risk, but equity, credit, inflation, currency, liquidity, fiscal and valuation risks remain. The central bank alters the distribution of risk across the financial system rather than removing it, which becomes extremely important when policy reverses.

QE is also bounded. It can reduce financial constraints; it cannot eliminate real economic constraints. Monetary policy can change financial prices, but it cannot manufacture real resources.

The transmission also depends on the state of the private sector. Heavily indebted households may use lower rates to repair balance sheets rather than to consume. Corporations with weak expected returns may not invest despite cheap financing. Capital-constrained banks may not expand lending. When valuations are already high, further easing may affect financial assets more than productive investment. The same accommodation can have very different effects depending on the structure of the economy.

QE is not necessarily permanent. When conditions change, central banks can reduce the size of their balance sheets, which is generally called quantitative tightening (QT). It can occur by allowing securities to mature, reducing reinvestment, selling assets or reducing lending facilities. The balance-sheet mechanism moves the other way: the central-bank balance sheet declines, securities remain in or return to private-sector portfolios, and reserves decline or other balance-sheet positions adjust.

QT is not simply QE backwards. Its effects depend on market conditions, the operating framework, reserve demand, banking-system liquidity, the pace of reduction, investor expectations, asset supply and term premia, and the financial system may respond very differently in tightening than during the original easing. The same chain works in reverse in outline: higher expected rates, higher yields and term premia, tighter financial conditions, lower valuations, higher debt-service costs and weaker borrowing and investment. The system becomes sensitive not only to the level of rates but to the change in the discount-rate regime. This is taken up in Lesson 10.

The deepest way to understand QE is through balance sheets. After a purchase, the central bank owns more securities, the private sector owns fewer of them, and the banking system holds more reserves. The critical question becomes what the private sector does with the assets it still owns.

The traditional macroeconomic question is how much money the central bank created. A more useful one is which assets the central bank removed from the private market, who sold them, what they received and where that capital went next. That is the logic of Follow the Assets.

When analysing any QE programme, seven questions keep the analysis on balance sheets rather than on headlines about the size of the programme.

Seven variables then give a fuller picture of monetary conditions than the policy rate alone.

The framework links monetary policy to asset prices. In a simplified cycle, low policy rates and QE lower term premia and trigger portfolio rebalancing, which raises demand for risk assets and their prices; higher prices improve balance sheets, ease financing and support credit and spending; higher incomes and profits then support asset prices further. When rates rise and QE is withdrawn the loop runs the other way: higher discount rates raise yields, portfolios are reallocated, valuations fall, financial conditions tighten, balance sheets weaken and borrowing and investment slow. The monetary cycle becomes a financial cycle.

In the end the lesson reduces to a chain: money, credit, banks, balance sheets, asset prices, financial conditions and the real economy. That chain is at the heart of modern financial capitalism.

• Ben Bernanke, 21st Century Monetary Policy, in the Core Canon; Hélène Rey, Dilemma not Trilemma, and BIS research on global liquidity and dollar funding, on the Dollar & International Monetary System shelf.
• Central-bank and BIS publications under Institutional Reading.