What Is Causing Inflation? And Can the Fed Lower It Without a Recession?
(9 min read)
(9 min read)
Inflation occurs when demand outweighs supply (demand-pull), for instance when the government increases its spending, or when production costs rise (cost-push), for instance when oil prices rise. When workers demand higher wages to match higher living costs and businesses raise prices in response to pass higher wage costs to consumers, a ‘wage-price spiral’ develops.
In truth, however, economists understand what is causing inflation poorly. Every 20 years or so, the dominant view changes:
These successive explanations suggest that the ongoing inflation surge will not correct itself. Instead, the Fed must tighten monetary policy, which is likely to cause a recession.
The 1950s witnessed the discovery of the Phillips curve, an empirical tradeoff between inflation, the rate of change of prices, and unemployment (Chart 1). The intuition behind the Phillips curve was that, when an economy nears full employment, firms have to pay higher wages to attract workers, and they pass these higher wages through to consumer prices. By the same token, a looser labour market kept wage costs, and thereby inflation, down.
The dominant view among economists was that the tradeoff was stable, and governments could therefore pick a preferred combination of inflation and unemployment.
The dissenters were the Monetarists, who claimed that the apparent relationship between inflation and unemployment was a money illusion. This illusion is a cognitive bias whereby people tend to mistake the face value of money for its purchasing power. And so higher inflation allowed for lower unemployment because workers incorrectly thought real wages (nominal wages minus inflation) were higher than they actually were.
The Monetarists believed that once workers figured out that real wages were lower than they thought, they would no longer work, and unemployment would rise back to its natural rate or the non-accelerating inflation rate of unemployment (NAIRU).
Chart 1 shows the relationship between inflation and unemployment was stable up to the late 1960s. In the 1970s, both inflation and unemployment rose, partly due to commodities price shocks. The breakdown of the Philips curve lent credibility to the Monetarists, who became the dominant voice.
Source: Macro Hive, Statistica
At its most basic, monetarism is expressed by the quantity theory of money:
where M is the stock of money, V is velocity (nominal income divided by money), P is the price level and T is real GDP. This is an accounting identity. That is, it is always true. And, on its own, it tells us nothing about the causal relationships between money, velocity, inflation or real GDP.
The Monetarists argued that the velocity of money was constant. They therefore thought an increase in money growth would translate into an increase in inflation, without a lasting impact on real GDP.
Indeed, Monetarists argued that the high inflation of the 1970s reflected excessive money growth.
For a while, the theory held. Inflation and money growth rose through the 1970s and fell through the 1980s (Chart 2). And it was this theory that inspired central banks’ stabilization policies of the 1980s, included those implemented by the Fed under Paul Volcker’s chairmanship.
Source: Macro Hive, Bloomberg
The low inflation the Monetarists predicted came to pass. But contrary to their predictions, the velocity of money was inconstant. It has been falling for the past 20 years. This led to the next vintage of inflation explanations: the expectations-augmented Phillips curve.
Central bankers dubbed the low inflation period from the mid-1980s to the Global Financial Crisis of 2008 ‘the great moderation’ – and promptly claimed credit for it.
They attributed the stability of inflation to well-anchored inflation expectations. That is, they thought that by conducting credible monetary policy, they had ensured consumers, businesses, and investors expected that prices would remain steady in the future. And that was important because central banks believed – on various theoretical grounds – that inflation expectations fed through to actual inflation.
In the expectations-augmented Phillips curve model, inflation dynamics comprise a series of deviations from a trend driven by long-term inflation expectations (Chart 3). As long as inflation expectations remain anchored, input price shocks, e.g., the current jumps in oil prices, only have a transitory impact on inflation. According to former Fed Chair Janet Yellen, ‘As a result, the central bank can “look through” such short-run inflationary disturbances in setting monetary policy, allowing it to focus on returning the economy to full employment without placing price stability at risk.’
Yellen was writing in 2015, when core inflation fell to 1.3%. She attributed this to ‘special factors likely to be transitory’. In fact, starting in the late 1990s, core PCE fell below 2%. It hit 1% in mid-1998. And it only recovered to above 2% shortly before the GFC. And post GFC, core PCE remained mostly stuck below 2%.
The Fed was concerned that inflation expectations could de-anchor, and that the US could slide into Japanese-style deflation. Therefore, it formally adopted Average Inflation Targeting (AIT) in August 2020. AIT is essentially a make-up strategy. The Fed allows inflation to overshoot its 2% target to make up for past undershooting.
In March 2021, inflation prints started to pick up. By February 2022, YoY core Personal Consumption Expenditures (PCE) inflation – a price index for capturing inflation across many consumer expenses – had jumped to 5.2%, from less than 2% a year before. Yet survey-based inflation expectations remained stable. Even the Fed’s own Index of Common Inflation Expectations (CIE) remained at 2.1%, the level prevailing in 2014, when core PCE was about 1.5%. The 1990s-2020 vintage of inflation explanations is ready for retirement.
Other than that we should treat models with a healthy dose of scepticism, I see four lessons we can learn from six decades of inflation explanations.
Phillips’ initial 1950s insight – resource pressures lead to inflation – was probably true. But the concept of resource pressure needs to be broadened to include both ‘demand-pull’ and ‘cost-push’ (supply-side) inflation.
First, let us look at the demand-pull side of the equation. The current inflation surge likely reflects that the monetary and, more importantly, fiscal response to the pandemic have created a positive demand shock. In response to the pandemic, the government implemented the largest increase in its deficit since World War II. The government budget included direct spending on items such as vaccines. But it also included very large transfers to households and businesses that boosted private demand. Also, because households were stuck at home, they increased their consumption of goods and decreased their consumption of services. This made for an unprecedented surge in demand for goods. This is the demand-pull.
Then there is the cost-push inflation from the pandemic. While the government was boosting its own and private sector demand, the pandemic led to pervasive global lockdowns. As a result, there were pervasive shortages of inputs, workers, and logistics services that forced businesses to cut down the amount of goods and services they usually supplied. These bottlenecks were starting to improve just when the Ukraine crisis happened. It has created a new negative supply shock through higher commodities prices and further disruptions of global logistics. All this contributes to cost-push inflation.
The scale of the shocks matter. In her 2015 speech, Yellen was referring to the post-1980s shocks that hit the US economy. These were much smaller than the current demand and supply shocks. The US 2020-21 fiscal easing was the largest since WWII; the supply disruptions induced by the pandemic and the Ukraine crisis are unprecedented in scope and depth. These make central bankers Great Moderation look as much good luck as good policies.
Inflation shocks are not always self-correcting, contrary to what the expectations-augmented Phillips curve implies. Even though fiscal policy has turned strongly restrictive and there were signs of improvements in supply bottlenecks before the Ukraine crisis US inflation has continued to accelerate. The shocks and resulting inflation are so large that Fed tightening will be needed to stabilize the economy.
Inflation expectations, while an exciting concept for economic modelists, likely have limited relevance in economies with stable and moderate inflation. In September last year, a senior Fed economist wrote ‘Economists and economic policymakers believe that households’ and firms’ expectations of future inflation are a key determinant of actual inflation. A review of the relevant theoretical and empirical literature suggests that this belief rests on extremely shaky foundations. Adhering to it uncritically could easily lead to serious policy errors.’ I am not sure the paper will do much to further the author’s career, but it sure hit the nail on the head!
The recent shocks to the US economy are likely too large to be self-correcting. Therefore, lowering inflation will require the Federal Reserve (the US’s central bank) to tighten their monetary policy by starting to raise interest rates. That would raise the cost of borrowing of businesses and households, lower demand for interest sensitive sectors of the economy (mainly real estate and consumer durables) and, the Fed hopes, slow inflation.
But here, the precedents are not encouraging. Whenever the Fed let itself fall behind the curve, as it currently has, a recession has ensued. And the longer the Fed has taken to tighten, the bigger the output and employment losses. At Macro Hive, we think the Fed has been too slow to respond to inflation, and we are anticipating a recession in 2023.
Inflation is the rate of change of prices. The US has two measures of inflation, the Consumer Prices Index (CPI) and the Personal Consumption Expenditures (PCE) Price Index. PCE inflation tends to be lower than CPI inflation.
Inflation happens when the rate of change of prices is positive , and deflation happens when it is negative. Slower inflation, and sometimes deflation, are usually associated with a recession.
Right now, inflation is the highest it has been since 1982. February PCE is 6.2% and core PCE, which excludes food and energy prices because they are volatile, is 5.2%.
Inflation occurs when the demand for goods and services is greater than the supply. It can be caused by excessive demand growth, for instance when the government increases its spending. Inflation can also be caused by a contraction in supply. For instance when oil prices go up, businesses that cannot afford the higher oil price have to close.
An oil shock is an example of cost-push inflation. Higher oil prices imply that the cost of business is going up, especially for businesses that consume a lot of it, for instance trucking. If truckers cannot raise the price of their services and pass on the higher oil costs to their customers, they may have to close.
The money supply is the sum of all financial assets that can be used as means of payments, mainly cash and bank deposits. The money supply can be saved or it can be used to purchase goods and services. If it is used for purchasing, it can contribute to demand-pull inflation.
Devaluation is when the value of a currency goes down. For instance, when the British pound’s value goes down, more pounds are necessary to buy a unit of foreign currency, for instance the Chinese renminbi. As a result, imports from China tend to become more expensive and can contribute to the UK cost-push inflation.