The 2% inflation target of most central banks has become the cornerstone of monetary policy during the last 30 years. How was that number chosen? If price stability is the main variable to optimize, why not target 0%? If the target is 2%, why does inflation regularly exceed it nowadays? Is a return to 2% even possible? Perhaps it's too late and that rate is already a relic of the past that is unattainable without devastating pain and hardship.

The 2% inflation target originated in New Zealand in 1990. Throughout the previous decades the country was going through a tough time - oil crises caused prices to surge, contributing to stagflation (high inflation combined with high unemployment). The Reserve Bank of New Zealand (RBNZ) was previously prioritizing production, trade, and full employment on a case-by-case basis. Because of this, the public's inflation expectations could also not be easily anchored. With inflation rampant, taming it soon became the RBNZ's primary objective. In 1990 the goal of keeping it between 0 and 2% was formalized. That range was not chosen in any rigorous way. It was the range at which previously prices had felt stable. Other countries soon followed with the explicit targeting of inflation. And that's how it has been for the last 30 or so years.

Central Bank Basics

A central bank (CB) controls the money supply of a country through an intricate system of reserves between itself and the commercial banks. Let's look at the Federal Reserve as an example. It has actually undergone big changes in the way it works. Prior to the 2008 crisis it operated in a scarce reserves regime, but the massive liquidity injections through quantitative easing made that old system obsolete, and since then it's operating in an ample reserves regime.

In the regime before 2008, banks were legally required to hold a percentage of deposits as reserves (fractional reserve banking). These reserves were kept low and scarce, forcing banks to rely on the federal funds market to borrow and lend reserves to meet legal requirements. The CB's primary policy tool were the Open Market Operations (OMOs): daily buying/selling of bonds to manipulate the quantity of reserves. The Fed had to precisely estimate and manage the supply of reserves to intersect the reserve demand curve at the desired Federal Funds Rate (FFR) target.

Post 2008 and currently, after the whole banking sector was flooded with liquidity, reserves are high, and banks are generally satiated. There are no reserve requirements. OMOs are used only periodically to ensure reserves remain ample. The main tools which determine the Federal Funds Rate (FFR) are the Interest on Reserve Balances (IORB) and the Discount rate. Here's a summary:

  • Like before, there is a market for overnight interbank federal funds. The average interest rate there is the FFR, targeted by the Fed. It influences the entire financial sector - bond yields, loans, indices, stocks - through various expectations and term premia.
  • The Fed sets the IORB - the interest on banks' money held in reserve accounts at it. A normal bank would not lend money in the Federal Funds market for less than it could earn risk-free from the Fed. Some lenders cannot earn IORB, so this does not apply to them.
  • The Fed sets the discount rate - the interest rate it requires on loans borrowed from it. It's usually set above the FFR, encouraging banks to borrow from the Federal Funds market instead of directly from the Fed.
  • Banks settle payments using Fed reserves. They cannot create them. The Fed controls the price of the settlement asset (reserves) that all banks are legally required to use. This monopoly power anchors the interbank overnight rate to the Fed’s desired level. That's the coercive element which keeps the system going.

In essence, in the ample reserves regime, the marginal value of additional reserves is approximately zero, so any market rates fall to the level of whatever return the central bank guarantees on reserve balances. That's why the IORB has become so significant as a policy tool. Note that, as before, newly created money is not backed up by any hard, scarce asset, only legal obligations.

The Cantillon Effect

By changing the IORB and the Discount rate the Fed can change the cost of reserves, which in turn affects the ex-nihilo money creation that happens in commercial banks whenever they give out loans. Through OMOs the CB can directly inject new money, boosting the money supply.

Yet the new money doesn't spread evenly. It creates winners and losers. Those who receive it before prices have risen, benefit the most. These are banks, primary dealers, institutional investors, and large corporations who are the direct counterparties for the Fed's asset purchases (OMOs) or who borrow directly from the interbank market.

Consumers, wage-earners, retirees, and those on fixed incomes receive the new money only after it has circulated through the financial system and the corporate sector. By the time it reaches them, asset prices and consumer goods prices have already been bid up, effectively eroding the purchasing power of their existing money and new income. So in effect there are two economies: one for asset-holders, enriched by monetary expansion, and one for wage-earners, crushed by the cost increases that follow. This is called the Cantillon effect.

Consider that the 2% inflation target is in terms of CPI. If inflation is 5%, the CB must tighten financial conditions enough to reduce aggregate demand and re-anchor expectations so that inflation returns toward 2%. This reduction will hurt everybody. Not only the households themselves, but also the institutional investors, pension funds, state governments that rely on continuous expansion only to stay solvent. Such a move can destroy a huge amount of wealth.

Sovereign Debt

It is common for governments to spend more than what they earn. Whether this constitutes good fiscal policy, or whether their earnings are obtained in a morally-acceptable way, are different questions. In the simplest case, suppose \(G_t\) is the government spending, \(T_t\) is taxes collected (net of any transfers), \(r\) is the real interest rate on debt, and \(B_t\) is the debt itself. Debt grows as

$$ B_t = (1 + r)B_{t-1} + \underbrace{(G_t - T_t)}_{\text{primary deficit}}. $$

Governments don't want to reduce spending \(G\). Instead, they want to increase it, for example by spending on social programs that benefit their electors. So a reduction in debt will likely have to come from increased taxes, by offloading the costs on the general taxpayer. Holding \(G\) fixed, and assuming that debt will have to be repaid:

  • A decrease in taxes now will have to be offset by a larger increase in taxes in the future. The more the government waits, or the higher the interest \(r\), the larger the increase will have to be.
  • To stabilize the debt, a government must eliminate the deficit.
  • To eliminate the deficit, if the \(r > 0\), we need \(T_t - G_t = rB_t\), requiring higher taxes forever.

How does huge government debt affect the economy and the spending decisions of households? One interesting hypothesis is Ricardian equivalence. It states that consumers are forward-looking and internalize the government's budget constraint when making consumption decisions. So if the government runs deficits for a long time, consumers will start to expect a future hike in taxes and will increase their saving now. This increased saving will offset the reduced public saving from the government and the effect of the large deficits will be relatively neutral.

Whether Ricardian equivalence holds is doubtful. For one, it assumes that future generations will be willing to pay potentially much higher taxes because of the deficits incurred through earlier generations. Beyond households, a high debt-to-GDP ratio implies more difficult borrowing and higher yield of government bonds, due to risk premia, which can also weaken the currency.

Lastly, if the CB increases interest rates, so that inflation can be brought back to 2%, the real debt will increase as well, potentially triggering a sovereign default. No government will allow the CB to do that, as it can lead to its own ruin.

Leverage and Wipeout

When central banks keep rates artificially near zero for extended periods, financial institutions fundamentally alter their business models, making them structurally dependent on easy credit. A small rate hike can then rapidly destabilize them. Here's a brief sketch of how it works.

Banking is the business of using short-term money to fund long-term assets to earn a profit. Suppose bank A wants to invest 100\$ in a long-term bond that pays a fixed 3% interest rate. To do so, bank A borrows 100\$ overnight in the Federal Funds market at a short-term rate of 0.5% and buys the bond. The bond is a fixed-income long-term asset, while the overnight loan is a short-term liability. However, now the loan is due tomorrow, while the first coupon from the bond (itself insufficient) will come a year from now. To repay the loan, bank A would have to either sell the bond prematurely (pointless), find funding from demand deposits or a new loan (what happens in reality), or sell assets to repay the loan (uncommon, sign of distress). Because the bank must perform this rollover every day to hold its long-term asset, it is continuously exposed to whatever the current overnight interest rate is. If the Fed raises rates to 5%, the bank immediately becomes unprofitable. This kind of maturity mismatch is an example of asset-liability mismatch.

This example, though illustrative, explains the underlying risk accurately. Isn't it shocking that banks work like that? They're highly leveraged institutions, typically funding 90% or more of their assets with debt. The entire system relies on the confidence that the bank can meet its obligations. When a bank's capital is threatened by rising rates, that confidence can vanish instantly, leading to the rapid deposit withdrawals known as a bank run.

Given all of the above, things look grim. There are three options:

  • A deflationary depression that liquidates the debt overhang, and likely the social order with it;
  • A financial repression that slowly confiscates wealth through negative real rates;
  • A restructuring of how we conceptualize monetary stability in a hyper-financialized economy.

The first option is politically unlikely and will cause massive economic damage. The second one is what's currently going on. The third one requires facing the fact that we've designed a system requiring constant monetary expansion to avoid collapse.