Quantity Theory of Money

The money supply is one of the key drivers of inflation rates, as suggested by the quantity theory of money. That is, the quantity theory of money postulates that the money supply has a positive (direct), proportional relationship with the inflation rate. Put differently, ceteris paribus, a 1 percent increase in growth of money supply would increase inflation rate by 1 percent. The magnitude of the relationship between the money supply and inflation may vary (depending on phases of the business cycle along with other factors), but the relationship between the two variables is well established in the literature.

Interest rates also play a role in driving inflation. Imbedded within the nominal interest rates is an inflation component. As described in the Fisher equation, the nominal interest rate is equal to the real inflation rate plus inflation.

Since interest rates are the cost of holding money, interest rates (nominal interest rates in particular) influence the demand for money. From the quantity theory of money we know that the money supply also influences prices and therefore inflation may have a two-way causal relationship with interest rates. Higher inflation expectations may therefore boost nominal interest rates because the FOMC would be expected to raise the fed funds rate to control inflation. Therefore, in theory, there is a link between money supply/demand, interest rates, and the inflation rate.

A Phillips Curve Framework

The money supply and interest rates alone, however, do not determine inflation. Other factors, such as aggregate demand, can affect the strength of the relationship between the monetary base, interest rates, and inflation. A practical example is the U.S. economy following the Great Recession. The FOMC increased the money supply significantly. Between late 2008 and mid-2014, the monetary base expanded by more than $2.5 trillion as the Federal Reserve embarked on three rounds of asset purchases, or quantitative easing. Over the same period, the fed funds target rate was left at 0 to 0.25 percent from December 2008 to December 2015. The inflation rate, however, continued to register significantly below the FOMC’s target of 2 percent. The average rate of inflation from 2009 to 2014 was just 1.4 percent, as measured by the PCE deflator.

The post–Great Recession environment runs contrary to the quantity theory of money as well as Milton Friedman’s statement that inflation is a monetary phenomenon. A possible justification of the broken link between the inflation and monetary variables is lack of aggregate demand. That is, following the 2007–2009 recession, the economy’s recovery was painfully slow. While the economy began to grow again in mid-2009, the pace was well below its historic average. In addition, slack in the labor market persisted, with full employment not being reached for another seven years. Demand remained weak as credit growth slowed and households and businesses deleveraged, leading to a slower velocity of money.

The impact of aggregate demand on inflation is typically explained through the Phillips curve framework. The Phillips curve suggests a relationship between the unemployment rate—a proxy for demand in the economy—and wages and inflation. The unemployment rate is inversely correlated with inflation. Greater demand drives lower unemployment, and as excess resources in the economy, in this case labor, are absorbed, prices move higher. This relationship between stronger demand and higher prices is referred to as demand-pull inflation. With more workers earning higher wages, aggregate spending increases, pulling up prices as “too much spending follows too few goods.” Therefore, at least in the short run, the effectiveness of money supply and interest rates as inflation drivers may depend on the aggregate demand.

Throughout history, the strength in the link between inflation and the unemployment rate has varied. After Arthur Phillips first published work on the link between unemployment and wages in the United Kingdom in 1958, economists Paul Samuelson and Robert Solow found a similar relationship in the United States between unemployment and inflation.¹ However, the link seemed to break down in the 1970s, when both inflation and unemployment in the United States soared (Figure 8.3).

The breakdown highlighted how the trade-off between lower unemployment and inflation was only a short-run phenomenon and excluded supply shocks. Policy makers could boost aggregate demand in the short run, leading to temporarily higher inflation, but actual inflation also depends on expected inflation. If economic agents expect higher inflation, then for a given level of unemployment, inflation becomes higher. The inclusion of inflation expectations is known as the augmented Phillips curve and can be written as follows:

where π is inflation, u – uⁿ is the difference between the actual and estimated natural rate of unemployment, and β measures how sensitive inflation is to unemployment. The term π_–1 illustrates that current inflation depends in part on recent inflation, as recent inflation affects inflation expectations and therefore price setting, wage demands, spending, and so on. Shocks, such as the oil price collapse in 2014–2016, also influence inflation and are represented by the term v.

More recently, the unemployment rate in the United States has fallen from a peak of 10 percent following the recession to less than 5 percent. Yet inflation has been little changed, leading some to question the efficacy of the Phillips curve. Stable and relatively low inflation expectations in recent years in part explain why actual inflation has remained so low in the Phillips curve framework. However, even when accounting for inflation expectations, the relationship between unemployment and wages/inflation has diminished (i.e., the slope of the Phillips curve has declined), although the relationship is still found to be present.²

Moreover, the relationship between unemployment and wages/inflation depends on the economy being at full employment. If the economy is not yet at full employment, then excess labor reduces the need for employers to raise wages. This leads to the idea of the nonaccelerating inflation rate of unemployment (NAIRU), also known as the natural rate of unemployment. Stronger demand would not be expected to drive inflation higher until unemployment has reached its natural rate. Estimates of the natural rate of unemployment vary over time as structural factors in the economy change (Figure 8.4). In recent years, Federal Reserve officials and the Congressional Budget Office estimate the natural rate of unemployment at around 4.8 percent. Through the first half of 2016, the unemployment rate averaged 4.9 percent, suggesting wage and broader inflationary pressure would only now begin to be building in the post-recession environment.

Cost-Push Inflation

In addition to demand, supply side shocks can also alter inflation, particularly in the short run. Sharp reductions in the supply common products can send price higher, raising overall inflation. This scenario is known as “cost-push” inflation. The most well-known example of this situation is the oil crisis of the 1970s. In retaliation for supporting Israel in the 1973 Arab-Israeli war, OPEC members placed an embargo on exports to the United States and other countries supporting Israel and also cut production. Between the middle of 1973 and the end of 1974, oil prices more than tripled. The rapid rise in energy costs sent overall inflation, as measured by the consumer price index (CPI), up from 3 percent in 1972 to 11 percent in 1974.

Cost-push inflation can also work the other way and lead to lower inflation. More recently, OPEC’s decision not to cut production to prop up prices amid rising North American supply sent shockwaves through the oil market. Oil prices fell around 75 percent from the middle of 2014 to early 2016. The lower cost of oil and related products pushed overall inflation lower, with the CPI briefly falling below zero. In addition to prices for energy goods falling, prices for transportation services also declined, pushing core goods prices lower, as it was cheaper to ship products.

Sometimes, past events (either demand-pull or cost-push) or expectations lead to inflationary pressures in an economy, which is referred to as built-in inflation. For example, a persistent demand-pull or cost-push scenario may increase inflation expectations and workers may push for higher wages in anticipation of higher inflation. In addition, producers may raise prices expecting higher input prices as well as higher wages.

Inflation Expectations

Recent inflation experiences can play an important role in expectations of future inflation. The inflation effects of short-term fluctuations in demand or supply previously discussed can shape how businesses and consumers think about future inflation. Following the 2014–2016 drop in oil prices, both market and consumer expectations for inflation fell. Given the significant role of energy prices on overall inflation, the decline in short-term inflation expectations reflected recent price movements. However, long-term inflation expectations (both market-based measures of compensation and consumer expectations) also fell to record lows (Figure 8.5). The drop in long-term expectations came even though oil prices could not fall forever and prices only needed to stabilize, let alone rise, for inflation—which, again, is the rate of change in prices, not a level—to move higher.

The drop in the public’s inflation expectations may have in part been driven by the Federal Reserve’s own forecasts. As mentioned earlier, the FOMC publishes its forecast for key economic variables, including inflation, regularly and market participants (financial markets in particular) pay close attention to those forecasts. Inflation forecast from the FOMC play a role in shaping the inflation expectations among economic agents. For more than two years, the FOMC repeatedly cut its inflation forecasts for the 2016 core PCE deflator (Figure 8.6). The persistent cuts to the Fed’s own forecasts of future inflation, along with the recency bias of lower realized inflation, likely weighed on the public’s perception of future inflation. A simple ordinary least squares (OLS) regression of core inflation based on the unemployment rate gap (measured by the unemployment rate minus its estimated natural rate) and long-term inflation expectations shows inflation expectations play a significantly larger role in determining core inflation.

Inflation Expectations and the Case of the Missing Deflation

Even as credit growth and broad demand have improved as the expansion has continued, inflation has remained stubbornly low. Yet this is not the first time in this cycle where inflation has been “stickier” than predicted by slack-based models. The rise in the unemployment rate to 10 percent and 4 percent decline in real gross domestic product (GDP) signaled a collapse in demand that should have also ushered in a collapse in prices, at least temporarily. However, core inflation slowed by only a little more than 1 percentage point and, despite the worst downturn since the Great Depression, remained around 1 percent. The mild weakening in inflation compared to what models predicted led economists on a search to explain the “missing deflation.”

The search has highlighted the importance of inflation expectations in determining realized inflation. While the unemployment rate more than doubled and hit nearly a 30-year high, long-term inflation expectations, measured by the University of Michigan’s Consumer Sentiment Survey, fell from an average of 3.1 percent in 2008 to 2.8 percent in 2010. Models including inflation expectations, and their relative stability in the downturn, better captured the behavior of inflation.³

Wages and Productivity in Inflation

As the expansion has proceeded and inflation has remained low, the modest rates of inflation now look to be altering consumer and business inflation expectations for the future. While the labor market has neared full employment and core inflation has improved, inflation expectations—both short and long term—began to slide from what were already low levels in late 2014.

The reduction in inflation expectations alters the path and intensity of future price and wage increases. Disappointing wage growth has been a common refrain since the recession. Explanations include the traditional slack arguments where, with the labor market only recently reaching levels that could be considered full employment, employers had little need to boost wages in order to attract or maintain workers. Research has also pointed to the stickiness of wages, where employees are reluctant to accept wage cuts. Therefore, even as slack rises, wages do not fall. In order for employers to adjust real wages, they therefore reduce the rate of future wage increases in what is referred to as pent-up wage cuts.⁴

However, the path of productivity plays an important role in the wage debate as well as its impact on inflation. Productivity is the ultimate driver of real wage gains as workers earn the marginal product of their labor. After an initial bump in 2009–2010 when demand began to rise again, yet employers were hesitant to hire, labor productivity slowed to a crawl, averaging just 0.5 percent per year from 2011 to 2016 (Figure 8.11). The weak rate of productivity growth along with low inflation was consistent with slow nominal wage growth.

If productivity remains weak, then diminishing slack in the labor market can more quickly lead to inflation as employers need to increase nominal wages faster than labor’s output. If workers are not producing more with each hour worked, raising topline revenues, employers cannot boost wages without cutting into margins or passing on higher labor costs through higher prices. The result is either a shift in income allocation from capital (profits) to labor, or inflation that offsets any nominal wage gains.

Recently, however, the link between labor costs and inflation has broken down more generally. The inability for employers to cut wages during the downturn (downward nominal wage rigidity) has been pointed to as one factor that kept inflation from falling as much as output-gap models would have suggested following the Great Recession. Yet the pass-through effects of higher wages and labor costs to inflation have weakened since the early 1980s, suggesting a diminished link between inflation and economic activity.⁵ The breakdown follows a declining share of income derived from labor earnings over the past few decades, weighed down by an aging population and falling labor force participation among prime-age workers (Figure 8.12).

Not Going It Alone: Low Inflation Globally

The United States is not alone in its low-inflation experience. Around the world, many other advanced economies have struggled to bring inflation back up to rates consistent with central banks targets. In the Eurozone, where the European Central Bank aims to keep inflation under 2 percent, core consumer price inflation has hovered around 1 percent in recent years, while headline inflation has been near zero (Figure 8.13). In Sweden, inflation has been below the Riksbank’s target since 2012, while inflation in Switzerland has been well below 2 percent since the financial crisis (Figure 8.14).

Japan has struggled most of all with low inflation. Prices have bordered on the brink between inflation and deflation for the past two decades, averaging just 0.1 percent per year. The inflation history of Japan highlights the challenges of persistently weak demand, exacerbated by unfavorable demographic trends. While Japan’s shrinking population makes the country’s demographic challenges more severe than any other major economy, other advanced economies, including the United States, face slower population growth and an aging population that portends slower growth and weaker inflation pressure.

The ubiquitous story of inflation running too low for the comfort of central banks follows the similarly common story of record monetary accommodation. Even as the Federal Reserve’s quantitative easing generated little help for inflation, it did not stop the European Central Bank from embarking on QE in early 2015 and for the Bank of Japan to further expand its balance sheet. QE was also undertaken with the aim to lower interest rates beyond central banks’ main policy rates. However, to further support inflation, a number of countries have lowered policy rates into negative territory, in the hopes of spurring credit growth and therefore demand in the economy. Less explicitly stated by central bankers is the currency effect of greater policy accommodation. As monetary policy eases, currencies depreciate amid capital outflows, all else being equal. The weaker currency can at least temporarily boost inflation via import prices. Yet, as evidenced with the inflation performance of the Eurozone, Sweden, and Japan, negative rates have done little to boost inflation.

The United States has been at the losing end of the currency/inflation dynamic. Even as the Federal Reserve has been slow to remove policy accommodation since the Great Recession, the relative strength of the U.S. economy has led to expectations that policy is more likely to tighten than ease, and is likely to tighten ahead of other major central banks. While the Fed has been talking about lifting rates, monetary policy in other advanced economies has become more accommodative or signaled that tighter policy would come at a later date than expected by the Fed. The divergence in policy and even the outlook for future policy caused the dollar to strengthen against a wide range of currencies. Between mid-2014 and the end of 2015, the broad trade-weighted dollar index strengthened more than 20 percent.

Lower import prices ensued. The strength in the dollar led to the effective cost of imported products falling as it took fewer dollars to purchase the same amount of foreign-denominated goods. Import prices for nonfuel items sank between autumn 2014 and the end of 2015 (Figure 8.15). Although the impact on overall consumer price inflation was limited since the majority of consumer spending is devoted to services, very few of which are imported, the stronger dollar contributed to weaker goods inflation. The CPI for goods excluding food and energy fell for the better part of 2013–2016, making it a drag on overall inflation (Figure 8.16).

The dollar is not the only factor, however, in weaker goods inflation. Globalization has opened up more markets, generating new sources of low-cost labor. Stronger productivity growth in manufacturing compared to services has also helped to reduce goods inflation in the United States and other economies.

Although inflation has not been as low as in advanced economies, developing countries have also experienced disinflation since the global financial crisis (Figure 8.17). At the forefront has been lower inflation in China. Consumer price inflation has slowed to around 2 percent after reaching 6 percent as recently as 2011. Overcapacity in the country’s industrial sector has led to the producer price index falling since 2012. The ongoing deflation in the “world’s factory” has contributed to lower goods inflation elsewhere in the world, via both cheaper prices for Chinese-made products, as well as tougher competition for foreign producers.

The disinflationary trends over the past few years have been more welcome in many emerging markets, as inflation remains far from deflation territory. However, the similar trends in both advanced and developing countries highlight the weakness in demand globally since the recession, evidenced by global growth remaining under its long-run trend since 2012. Amid the slowdown, lower commodity prices have been a common theme across the world, but have played a larger role in developing nations since food and energy make up a larger share of consumer purchases.