Central banks and big investors look at market prices every day to guess where inflation is going. But these prices do not always reflect what people truly expect.
In a recent study that I have co-authored with Robert Czech, Sitong Ding and Ricardo Reis, we use detailed new data on millions of trades to show that the market is split between different types of investors (Bahaj et al, 2025). The main message is that market prices often exaggerate how much people worry about inflation, especially during big crises.
A tale of 25 million trades
We study the UK market for ‘inflation swaps’ – contracts with which people trade insurance against rising prices. By looking at over 25 million individual trades, we find that the market is divided into two distinct parts. Pension funds usually buy long-term protection for ten years or more to cover their future payments to retired people. At the same time, hedge funds trade over much shorter periods, usually three years or less.
Banks act as the bridge between these groups. They are not just neutral middlemen, and often take on large risks. For example, the surprise jump in UK inflation between 2021 and 2023 is likely to have cost banks around $15 billion. We also find that banks ‘put their money where their mouth is’: their actual trades match the inflation guesses that they give in private surveys.
Implications for monetary policy
Our research matters because monetary policy-makers use these market prices to decide if they need to change interest rates up or down to achieve their objective of low and stable levels of inflation. We find that market prices often move because of market frictions – such as bank regulations or a lack of ready cash – rather than real news about the economy.
During both the Covid-19 pandemic and the energy crisis of 2021-23, market prices made it look as if people were concerned about a permanent shift in inflation. But our measures show that true expectations were much more stable.
In fact, when the market price for long-term inflation rises by 100 points, the underlying real expectation usually ‘only’ goes up by about 88 points. In short-term markets, the signal is even weaker. This suggests that policy-makers might react too strongly to market swings that do not reflect the reality of expectations about prices.
A new approach for central banks
Central banks should stop taking market prices at face value. Instead, they can use the methods that we present in our research to filter out market noise and see the true signal. This would help them to make better decisions about interest rates without being distracted by temporary market quirks.
Future research should also look at why it is harder for banks to provide protection for short-term trades than for long-term ones. Understanding these hurdles will help to make financial markets more stable and predictable for everyone.
