Costas Milas and Georgios Papapanagiotou looked at a number of variables to forecast inflation, GDP growth and unemployment in the UK. They compared these forecasts with those of the Bank of England and found that the BoE consistently overestimates unemployment, with strong implications for inflation. Based on their findings, they advise the Bank to start paying attention again to the money circulating in the economy.
What we have learned since the mid-2000s is that economic/financial shocks come big in size, including, in the words of David Smith, the economics editor of the Sunday Times, “the four economic horsemen of the apocalypse”, namely, the global financial crisis, Brexit, the pandemic and the Russian invasion of Ukraine. Add to this the ongoing war in Gaza between Israel and Hamas, which has increased geopolitical risk even further, to conclude that empirical modelling and forecasting in the presence of the above shocks has been extremely challenging.
In 2023, Bank of England governor Andrew Bailey recognised that the UK’s central bank has “very big lessons to learn” about the conduct of monetary policy in its battle to bring down inflation when large economic shocks occur. He thus tasked Ben Bernanke, former chair of the US Federal Reserve and winner of the 2022 Nobel Prize in Economics, with a review of the BoE’s models. In his report, Bernanke urged the Bank to “deal with uncertainty and structural change” in its forecasting analysis. He also noted that a revamped forecasting framework should pay attention to supply-chain disruptions.
In a new research paper, we take the advice of Ben Bernanke on board. We employ quantitative models with time-varying coefficients. These models look at financial variables (such as long-term interest rates, exchange rates or stock market indices), economic variables (such as wage pressures or economic policy uncertainty) and foreign variables (such as geopolitical risk, the infectious disease market volatility tracker, global supply chain pressures and oil prices).
The models allow these variables to impact the UK economy with effects that change over time. But all depends on whether the economic policies pursued by successive UK governments are detailed enough or carry a substantial element of economic surprise, such as the so-called “unfunded” tax cuts pursued by former Prime Minister Liz Truss in September 2022. Among all these variables mentioned above, geopolitical risk and oil price movements have become extremely relevant since Russia’s invasion of Ukraine and the Hamas-Israel war in Gaza.
In our paper, we employ these types of models to provide forecasts of GDP growth, inflation and the unemployment rate. We then compare our forecasts with those provided by the BoE. To our (slight?) surprise, and given the well-known criticism the BoE has recently faced regarding its forecasting failures, we find that the Bank’s own GDP growth forecasts are relatively more accurate than those of our more sophisticated models. (Their own inflation forecasts slightly less so.)
On the other hand, the BoE’s unemployment forecasts are much less accurate relative to those produced by our models. More to the point, the BoE has the habit of consistently overestimating actual unemployment. This very finding has important implications for inflation. A higher unemployment rate reduces wage demands made by workers because rising unemployment adds to their worries about the fate of their employment prospects. This suggests the BoE will most likely see weaker wage pressures on inflation than in reality, therefore producing inflation forecasts that are less accurate than intended!
We also find that a model of UK inflation that allows forecasts to depend on today’s inflation figure, global supply pressures and pandemic effects is largely able to forecast inflation better than the BoE. The attractiveness of this model is that it ignores what level of interest rates BoE’s policymakers will decide in the future!
The model is attractive because it bypasses the possible problem of forecasting based on what markets believe about the future path of UK interest rates. Financial markets have been wrong in the past about their interest rate predictions. Up to the time of writing, the BoE uses as “input” market expectations of interest rates to predict inflation, GDP growth and unemployment.
Wrong interest rate predictions made by the markets will invalidate the BoE’s forecast. On the other hand, we offer a forecasting model of inflation (GDP growth and unemployment) that, if used by the Bank’s policymakers, can provide relatively accurate forecasts without having to worry about market expectations of interest rates, whether they turn out to be correct or wrong!
We also find that money growth makes a significant comeback towards forecasting inflation. The idea here is that the more money circulates in the economy, the higher the consumption of goods and services, therefore strengthening GDP growth but also inflation. Notice that it is “fashionable” nowadays to ignore the inflationary impact of money since it has a poor record of forecasting inflation at least before the COVID pandemic. Our paper shows that central banks should start paying attention to money again!
To illustrate, Figure 1 plots, from 2004Q1 onwards, actual inflation together with divisia M4 annual growth four quarters earlier, and the so-called time-varying impact of money on UK inflation based on our work. Divisia M4 is the broadest measure of money in the economy, which weights different forms of money according to their likelihood of being spent. Notes and coins have a higher weight than money held in mutual funds, for example.
Figure 1. The impact of money on UK inflation in sample, 2004-2024
Source: Milas and Papapanagiotou (2024).
What do we notice?
Between 2004 and early 2009, the impact of money on inflation rose while inflation went up. In other words, the more money circulates in the economy when prices are already on the rise, the higher the impact of money on inflation. Put simply, more money throws fuel into the inflation “fire”.
Check what happened from 2020 onwards. Same story as above! The money impact rose again when past growth rates of divisia M4 were remarkably strong. More to the point, between the first quarter of 2020 (when additional quantitative easing of £200bn was authorised by the BoE) and the fourth quarter of 2022 (when inflation reached its peak rate of 10.7 per cent), UK inflation averaged 4.16 per cent per annum.
Our empirical findings suggest that divisia M4 growth contributed to this inflation figure by an annual average of approximately 0.31 percentage points. Most of us will agree that this very impact of money on inflation is not quantitatively much. However, from Figure 2, one can also see that our model with Divisia M4 money predicts inflation better than the BoE’s models since the pandemic. In other words, although money might have a small inflationary impact, it crucially predicts future inflation quite well! Therefore, the BoE should pay (more) attention to money when forecasting the UK economy!
Figure 2. Actual inflation and one-quarter ahead forecasts from the BoE and a model with divisia M4
Source: Milas and Papapanagiotou (2024)
Sign up for our weekly newsletter here.
- This blog post is based on On the (Un)Foreseeable Path of the UK Economy Through Uncertain Times.
- The post represents the views of the author(s), not the position of LSE Business Review or the London School of Economics and Political Science.
- Featured image provided by Shutterstock
- When you leave a comment, you’re agreeing to our Comment Policy.
Print Friendly
