Weakness in US core inflation has been in the spotlight recently. Despite the economy operating above potential, with a tight labour market, inflation has stayed subdued. One-off explanations like policy and methodological changes might partly account for this, but weak inflation is not only a US phenomenon. Most advanced economies, and some emerging countries, are witnessing lower inflation rates relative to similar points in previous cycles. Moreover, this is not solely a post-crisis phenomenon - inflation has been steadily coming down since the 1980s, with a significant downshift in the 1990s and another one over the last few years.

A simple exercise (involving some econometric modelling!) looking into the main drivers of inflation over time reveals some interesting trends. Firstly, the impact of the output gap (a measure indicating where the economy is operating relative to its potential) on core inflation in the US, UK and euro area has fallen over the last few decades, and especially since the late 1990s. In other words, inflation is not as sensitive to tighter labour markets or capacity constraints as it used to be. In economic terms, the much talked about Phillips curve, which depicts the normally inverse relationship between unemployment and inflation, is flatter, especially since the crisis, which is in line with some recent evidence. Interestingly, in the post-crisis period the UK output gap has become completely irrelevant in driving UK inflation.

Secondly, the impact of inflation expectations is also changing - though they are still an important determinant of inflation overall. In the US, the effect of inflation expectations has become less volatile and more consistent since the late 1990s, presumably as many central banks started shifting towards inflation targeting at the time, anchoring expectations. Interestingly, euro area core inflation has become more sensitive to inflation expectations since the sovereign debt crisis, arguably as a result of the European Central Bank’s extreme policy measures, as well as ECB President Mario Draghi’s increased focus on inflation expectations since then.

Thirdly, while inflation is a persistent process (whereby past inflation is a decent predictor of current and future inflation), the degree of its persistence has declined, particularly in the US and euro area since the financial crisis. Euro area inflation persistence stands out in terms of importance for determining current inflation relative to other components.

Finally, external determinants of inflation such as import prices are significant, but in line with more domestic drivers, have also become less important over the past few years. The currently low sensitivity of US inflation to import prices, for example, stands out relative to the mid-1990s and mid- 2000s.

While this offers some insights into the importance of inflation determinants over time, it does not explain what is behind these findings! No single factor (be it globalisation, technological progress, demographics, the internet, change in consumer and corporate behaviour, or policy) can fully explain the weakness in inflation we are observing today. But together, they probably do account for the lower sensitivity of inflation to traditional drivers and for the modestly disinflationary environment of the past few years.

This exercise is not exhaustive and its findings should not be over-interpreted, but there are some worthwhile conclusions. To begin with, given the lower sensitivity of inflation to its fundamental determinants, including output gaps, inflation expectations and inflation persistency, it is unlikely that we will see inflation accelerating or decelerating fast from current levels in the near-term. ‘Low and stable’ really is here for longer. In addition, it would take a large shock - perhaps a sudden adjustment to labour supply or a sharp change in inflation expectations - for inflation to break out of the current range meaningfully. This is not entirely impossible in the current environment; perhaps Brexit or Trump could provide a catalyst for that.

For policymakers, this complicates an already enormous challenge. The economy has to run above potential, with larger output gaps than we are currently seeing, for inflation to see a sustainable upshift, and inflation expectations may have to jump up (and stay higher). In other words, to raise inflation and meet their direct mandate, central banks have to stay ‘behind the curve’ for longer than their traditional models and policy rules suggest. This is not the strategy followed by the US Federal Reserve currently, but perhaps the ECB has a chance, given the extreme level of accommodation still in place? Of course, falling behind the curve would increase the risks of financial instability even further, which central banks are acknowledging but struggling to address. This is a tough trade-off. Damned if you do, damned if you don’t? I would not want to be a central banker right now!