- Why Is Inflation Low Globally?
- What drives inflation?
- What is new since the global financial crisis?
- Non-OECD countries and non-crisis trends
- How do economists try to predict inflation?
- Full employment
- Unemployment and Inflation Predictions
- House Prices and Future Inflation
- Money Supply and Inflation
- Hysteresis – What Happened in the Past?
- Difficulties of predicting inflation
Why Is Inflation Low Globally?
The world economies have mostly recovered from the 2008 financial crisis. In the United States, United Kingdom, and Germany, for example, the unemployment rate is now below 4%, lower than it has been in decades. Tight labor markets are usually a symptom of a healthy economy and thus a rising demand for goods and services.
To satisfy such demand, businesses usually raise prices—the basic mechanism underlying the standard economic relationship known as the Phillips curve. Yet, 10 years after the financial crisis, inflation has held remarkably steady.
Some researchers therefore argue that the Phillips curve is no longer a useful descriptor of inflation dynamics (Coibion and Gorodnichenko 2015).
In this Economic Letter, we investigate whether the financial crisis has changed the long-standing inner workings of the Phillips curve. We extend the analysis in our previous Economic Letter (Jordà et al. 2019) to include developing economies and analyze three components of the Phillips curve to assess where the links appear to be broken.
Our analysis suggests that news of the death of the Phillips curve in developed economies appears premature. Fluctuations in labor market conditions have been largely offset with appropriate interest rate changes by central banks. Under such conditions, the influence of past inflation has faded, and expectations for future inflation have gravitated toward the central bank’s stated target.
On the surface, inflation appears stable at all levels of labor market slack, and the Phillips curve link appears broken. Underneath, however, the Phillips curve could still be at work. The inflation dynamics that we observe could also be explained by a central bank that successfully offsets fluctuations in slack to keep inflation at the target.
In less developed economies, central banks often operate under constraints that prevent full monetary policy offsets, which has given a bigger role to feedback from past inflation. That said, inflation has been trending down for the past two decades across developed and developing economies a.
The inevitable conclusion is that there are global forces putting downward pressure on inflation, and it is not just the result of better monetary policy.
What drives inflation?
To assess how well the Phillips curve explains inflation, we treat the financial crisis as a quasi-natural experiment. Because the crisis was mostly unexpected, we can use the time before the crisis as the control or baseline for the Phillips curve relationship to examine what happened after the crisis.
We estimate a Phillips curve model that explains inflation as a function of three components. First, we measure the demand-pull factors, using slack in the labor market.
Specifically, we use the unemployment gap, which is the gap between the unemployment rate and its natural rate, or the rate at which prices would remain stable.
The unemployment gap is a common proxy measure for aggregate demand conditions because higher demand usually means more hiring.
Second, we use feedback from past inflation, which we measure with the headline consumer price index (CPI) inflation. This acknowledges that prices tend to adjust slowly, so where inflation has been can influence where it is headed. That is, businesses take some of their pricing cues from previous periods, therefore making inflation persistent.
Third, we include expectations of future inflation survey data for the United States and data provided by the Organisation for Economic Co-operation and Development (OECD) for all other countries.
If the central bank can credibly commit to an inflation target, then businesses will choose to price their products in a manner consistent with the stated target.
In such a setting, fluctuations in the unemployment gap and past deviations of inflation from its target have a much more muted effect on actual inflation. Credibility is therefore a precious commodity.
In the decade since the global financial crisis, the relationship between the unemployment gap and inflation appears to have weakened, as shown in Figure 1.
Each point in Figure 1 represents the average of the unemployment gap, along the horizontal axis, and headline CPI inflation, on the vertical axis, across all OECD countries for a particular quarter.
In the two decades before the crisis, there was a clear negative relationship between unemployment and inflation (yellow and green lines), such that when unemployment was high, inflation was low, and vice versa.
The current decade (red line) shows that relationship has all but disappeared. Even with the unemployment gap below zero—meaning that on average the unemployment rate is lower than its natural rate—inflation averages around or below 2%.
Phillips curve across OECD countries by decade
However, Figure 1 provides only a simple correlation. Our next step is to formally estimate how each component of the Phillips curve—slack, persistence, and expectations—contributes to inflation, and how those contributions changed around the decade before and after the global financial crisis.
What is new since the global financial crisis?
We move beyond simple correlations and formally estimate the contribution of each of the three components of the Phillips curve: the unemployment gap, to measure slack; past inflation, to measure persistence; and inflation expectations.
We also remove variation in inflation due to fluctuations in oil prices; this accounts for the fact that oil prices represent a classic supply factor that is outside the control of the monetary authority. We focus first on OECD economies that experienced the financial crisis.
Figure 2 summarizes estimates for each of these components over two periods, before and after the financial crisis.
Contributors to Phillips curve changes: OECD countries
The first pair of bars in the figure shows the slack component. The bars are negative because more slack means lower inflation, thus subtracting from the total.
Although the contribution of slack has dissipated since the crisis, it is clear that slack played a small role in explaining inflation dynamics before the crisis.
Given the magnitude of our estimates, even large values of the unemployment gap would have only a small effect on total inflation.
The second and third pairs of bars refer to persistence and expectations components, respectively. Since the crisis, the estimate of the persistence term has declined by as much as the estimate of the expectations term has increased.
Less persistence means that a perturbation to inflation today feeds into tomorrow’s inflation to a lesser degree. The increase in the estimate of the expectations term means that consumers are ly to dismiss such perturbations as transitory.
In our previous Letter, we documented a similar pattern in the United States and argued that well-anchored expectations are a natural consequence of credible monetary policy.
Non-OECD countries and non-crisis trends
Even before the financial crisis, several other global trends were taking shape that probably affected global inflation. As a starting point for assessing the influence of these trends, Figure 3 displays average inflation across countries since 1998, divided into OECD developed economies and a sample of 23 non-OECD developing economies.
Average consumer price index inflation
Figure 3 indicates that inflation for developing and developed economies has gradually converged over the years. By the end of the sample, average inflation in both groups was virtually the same.
In part, the decline observed in developing economies may reflect increased credibility of central bank policies.
However, both groups of economies share a common trend in inflation, suggesting that global factors may be keeping inflation at bay everywhere.
To better understand what might be happening globally, we repeat our estimation of the Phillips curve with the three components using the sample of 23 non-OECD developing economies in the decade before and after the global financial crisis. Figure 4 summarizes the results.
Contributors to Phillips curve changes: Non-OECD countries
The first pair of bars shows that the unemployment gap exerted as little influence on inflation in developing economies as it did in developed economies. The persistence and expectations terms are more interesting.
In contrast with what happened in developed economies, the persistence term was much larger and, if anything, has increased since the crisis. That is, perturbations in inflation today tend to play a larger role in how inflation will develop in the future.
Also, the effects of expectations in developing economies have diminished to a larger degree relative to their developed peers as well as relative to the decade leading up to the crisis.
Since the three components of inflation reflect different experiences among developed and developing countries after the crisis, those elements cannot explain the global decline in inflation being low globally.
Rather, the answer appears to lie in some other common underlying factors, which could be related to increasing trade openness, global supply chains, and greater capital and investment flows across countries.
Such factors began reducing costs of production and investments and putting downward pressure on prices around the globe even before the crisis (International Monetary Fund 2006).
The global financial crisis upended our understanding of inflation dynamics, particularly when viewed through the lens of the Phillips curve. Low inflation has persisted despite very low unemployment in developed countries.
Yet something similar has taken place in some non-OECD developing countries: despite the fact that nearly all of these countries have escaped the financial crisis, inflation has remained low there as well. Interestingly, for developing economies, the role of past inflation in explaining current inflation remains dominant, in contrast to the pattern among more advanced economies.
However, because all countries have experienced a similar decline in inflation, other global factors must have played an important role in recent subdued inflation.
Òscar Jordà is vice president in the Economic Research Department of the Federal Reserve Bank of San Francisco.
Chitra Marti is a research associate in the Economic Research Department of the Federal Reserve Bank of San Francisco.
Fernanda Nechio is deputy governor in International Affairs and Corporate Risk Management with the Central Bank of Brazil.
Eric Tallman is a research associate in the Economic Research Department of the Federal Reserve Bank of San Francisco.
The views expressed do not necessarily reflect the views of the Central Bank of Brazil, the Federal Reserve Bank of San Francisco, or the Federal Reserve System.
Coibion, Olivier, and Yuriy Gorodnichenko. 2015. “Is the Phillips Curve Alive and Well After All? Inflation Expectations and the Missing Deflation.” American Economic Journal: Macroeconomics 7(1), pp. 197–232.
International Monetary Fund. 2006. “How Has Globalization Affected Inflation?” Chapter III in World Economic Outlook, April.
Jordà, Òscar, Chitra Marti, Fernanda Nechio, and Eric Tallman. 2019. “Inflation: Stress-Testing the Phillips Curve.” FRBSF Economic Letter 2019-05 (February 11).
Opinions expressed in FRBSF Economic Letter do not necessarily reflect the views of the management of the Federal Reserve Bank of San Francisco or of the Board of Governors of the Federal Reserve System. This publication is edited by Anita Todd with the assistance of Karen Barnes. Permission to reprint must be obtained in writing.
How do economists try to predict inflation?
Readers Question: How does the MPC predict future inflation?
Inflation is caused by a mixture of demand-pull and cost-push factors.
Therefore, the MPC will look at many statistics which give an indication of whether the economy is reaching full employment and causing inflationary pressures.
This will include rate of economic growth, unemployment and the amount of spare capacity (output gap) in the economy. Inflation could also come from supply-side factors such as rising oil prices and rising wages.
Source B of E
This shows inflation forecasts for the next three years. The dark red shows most ly inflation rate, The lighter red shows the range of possible inflation forecasts.
This shows that forecasting inflation is far from an exact science. There are many factors affecting inflation and many of these are uncertain.
For example, an unexpected shock could cause inflationary pressures to rise (e.g. higher oil prices)
A key factor in predicting inflation is the amount of spare capacity and the rate of economic growth.
Suppose an economy, such as the UK, has a long-run trend rate of 2.5%. This means growth of 2.5% or less is unly to cause inflation. If however, growth is above – e.g. 3 -4 % then the economy will quickly approach full capacity and therefore inflation is ly to occur.
Inflationary pressure from rising AD. For example, in the late 1980s, the UK had growth of 4% a year, but this caused rising inflation.
UK Economic Growth
The long-run trend rate of economic growth in the UK is roughly 2.5%. If growth is below 2.5% (or the equivalent of 0.6% a quarter) then demand-pull inflation will be very low. This shows that in the past four quarters, the UK is currently experiencing below-trend rate of growth causing low inflationary pressures.
A slowdown in global growth also suggests that global inflationary pressures will continue to remain muted. With lower global growth, there will be lower demand for exports. Also, lower global growth will tend to cause lower prices for commodities, leading to less cost-push inflation.
Unemployment and Inflation Predictions
Some economists believe there is a trade-off between unemployment and inflation. If unemployment falls it could be a sign that inflationary pressures will increase.
A fall in unemployment typically causes inflationary pressures. As unemployment falls and job vacancies rise, usually this leads to wage-pull inflation – workers can demand higher wages.
However, the UK has an unusual job market – unemployment is low but wage growth is also low.
This has been a sustained reduction in the natural rate (structural unemployment) Workers don’t have much bargaining power and there is a substantial amount of under-employment.
House Prices and Future Inflation
It is often assumed that house price inflation will cause actual inflation. There is a good economic reason for this. If house prices are rising it creates a wealth effect.
Rising wealth encourages consumer spending (equity withdrawal and higher confidence) this spending can then cause inflation. However, it doesn’t necessarily cause inflation.
In the early 2000s, the UK had house price inflation of over 20%, but it didn’t cause actual inflation. There are many factors that affect inflation, not just house prices.
Money Supply and Inflation
The quantity theory of money states that increased money supply will lead to inflation.
This is because of the relationship between money supply and inflation, shown in the equation MV=PT where V and T are independent of the Money Supply.
However, in practise empirical evidence has shown that increased money supply doesn’t necessarily cause inflation, as there are other factors determining money supply and inflation.
Hysteresis – What Happened in the Past?
Sometimes the best way to predict inflation is to look at what happened in the past
It is argued that if you want the best prediction for inflation, ignore all the economist’s predictions and just state what happened last year. Of course, inflation does change from year to year, but it shows the difficulty of predicting inflation that it is often best to just use last years data.
However, there is an important point here and that is the role of expectations. If inflation is low, people will expect low inflation in the next year, workers will not demand big pay rises, firms will not try to increase prices.
Therefore, low inflation becomes easier to maintain. If inflation is high then people will be expecting inflation in the next year.
Therefore, it becomes difficult to remove inflation from the system (without pain a recession)
Supply-Side Shocks and Inflation
When predicting inflation you can never take into account unexpected supply-side shocks.
For example, a rapid increase in oil price inflation would cause a significant rise in inflation, as they did in the 1970s and 2008.
In 2020, recent events such as fears over the virus have caused a substantial fall in oil prices and commodities, which will make economists revise down their inflation forecasts.
Sustained real wage growth can cause inflation due to higher costs for firms – rising demand. This shows that since 2009, real wages have been relatively stagnant – so wage push inflation is still muted.
Other factors the MPC may look includ
- Exchange rate. A rapid devaluation in the exchange rate is ly to cause inflation from higher imported inflation. however, to complicate matters, this kind of inflation is ly to be temporary. The spikes in inflation in 2008 and 2011 were cost-push inflation.
- Balance of Payments – A deterioration in the current account may indicate an unbalanced economy and inflationary pressures leading to higher demand for imports.
- Consumer Confidence
- Level of Investment and business confidence
- Amount of spare capacity surveys.
- Input prices – prices of raw commodities.
Difficulties of predicting inflation
Long-term economic forecasting can be very difficult. A well known joke by John Kenneth Galbraith is:
“The only function of economic forecasting is to make astrology look respectable.”
Some difficulties include
- Data can change in importance. Falling unemployment used to be a good guide to predicting inflation, but recently the labour market has changed – with low wage growth despite falling employment
- Unexpected shocks in the global economy.
- Global economies are too complex to include all factors in models.
- Track record of predicting inflation is not particularly good. Forecasters tend to go for conservative predictions and miss the big swings.
- Hard to know impact of monetary policy. If interest rates rise, will consumers respond by cutting spending.