Federico Pessina, Maren Froemel and Ivan Yotzov
Understanding inflation expectations is key for monetary policy makers and has been central to the policy debate in recent years. We use data from the Decision Maker Panel (DMP) – an economy-wide UK business survey – to analyse businesses’ expectations about aggregate CPI inflation, and the relationship with their own-price expectations. On average, firms are attentive to current inflation rates, but larger and more productive firms report more accurate perceptions and expectations. In recent years, both one-year and three-year CPI expectations have become more sensitive to inflation perceptions, and three-year CPI expectations have also become more sensitive to one-year expectations. Finally, aggregate dynamics matter for firms’ decisions: CPI expectations are correlated with firms’ own-price expectations and more so for more productive firms.
The Decision Maker Panel (DMP)
The DMP is a monthly online survey of UK businesses with ten or more employees. Launched in 2016, it is run by the Bank of England, in collaboration with King’s College London and the University of Nottingham. In December 2024 the survey reached its one-hundredth wave. Firms are regularly asked about the annual growth in their output prices, as well as about year-ahead own-price expectations. In addition, since May 2022, firms have reported their current CPI inflation perceptions, one-year CPI expectations, and three-year CPI expectations. By January 2026, the survey had received 33,000 responses on CPI perceptions and expectations from over 6,000 unique firms.
The dynamics of firms’ CPI perceptions and expectations
Chart 1: Average perceptions track CPI closely, while near-term expectations have declined significantly since 2022
Note: CPI expectations are plotted at the survey date and are not time-shifted forward.
How have firms’ perceptions and expectations of CPI evolved since 2022? Chart 1 shows that firms’ perceptions of current CPI (red line) have tracked CPI inflation closely (grey bars), with an average absolute perception error of only 0.31 percentage points (pps). In contrast, CPI inflation expectations have been less accurate. For example, one-year CPI expectations (green line) have an average absolute forecast error of 1.55 pps over this period. These near-term expectations also declined more gradually than CPI inflation over 2023. Year-ahead CPI inflation expectations were on average 4.9% in June 2023, whereas a year later CPI inflation had fallen to 2%. Gautier et al (2025) noted a similar pattern, using survey data from French firms. Finally, three-year CPI inflation expectations have been more stable since 2022, ranging from 2.5% to 4.5%. In mid-2024, perceptions and expectations converged at around 3%. Since then, the series have gradually diverged, jointly with higher CPI inflation.
Chart 2: Disagreement between firms in perceptions and expectations have fallen
Chart 2 depicts how the standard deviation of perceptions and expectations (a commonly used proxy for disagreement between firms) varied over time. Dispersion in perceptions and expectations is positively correlated with the inflation rate. Interestingly, between 2022 and 2024, dispersion for one-year CPI was higher than for three-year CPI inflation. Since 2024, firms’ beliefs around one-year and three-year CPI have shown a similar level of dispersion, whereas disagreement around current CPI inflation has fallen further. Chart 2 also provides indirect evidence on expectation errors. While the standard deviation of beliefs declined as inflation fell, the dispersion relative to the mean fell less and, in some cases, even increased. This measure reflects that an absolute error – say, 1 pp – is relatively larger and more meaningful at lower levels of inflation.
Firm-level determinants of perception and expectation errors
Past research shows that firms’ own-price forecast errors are negatively correlated with profitability and total factor productivity (TFP). The ability to correctly perceive current conditions and predict future ones may have important implications for long-term performance, and performance may in turn reinforce accuracy. We begin by analysing whether CPI perception and expectation errors are correlated with key firm characteristics, such as productivity and size.
Chart 3: Larger and more productive firms make smaller absolute forecast errors
Notes: Controls: industry fixed effects and time fixed effects. Includes 95% C.I. Coefficients from joint regression including all firm determinants. Dependent variables standardised to mean zero and unit variance.
Chart 3 shows how two key firm characteristics – labour productivity and number of employees – correlate with CPI perception, CPI expectation and own-price expectations errors. The coefficients are based on regressions with industry and time fixed effects, and we standardise the dependent variables to have zero mean and unit variance for comparability. Higher firm productivity (left panel) is associated with significantly smaller perception and one-year CPI expectation errors. The relationship with three-year expectation errors is also negative, although less precisely estimated, likely due to the smaller sample for these series. The relationship with the number of employees (right panel) shows similar patterns, with significant effects across all perception and expectation error measures. Furthermore, we find a stable relationship between these firm characteristics and expectations over time. Overall, we conclude that firm productivity and size are robust determinants of firm perception and expectation errors, across multiple definitions and horizons.
Firm-level relationship between CPI perceptions and expectations
Chart 4: Perceptions of high current inflation are associated with higher expectations of future inflation
Notes: Controls: firm fixed effects, time fixed effects and CPI inflation.
How sensitive are inflation expectations to current inflation? We analyse the relationship between firm-level perceptions and expectations, how it varies over time and across firm characteristics. While related to work on the impact of aggregate inflation data releases, our analysis uses firm-level variation in both perceptions and expectations. Chart 4 presents the coefficients of a regression of CPI perceptions on one-year and three-year expectations, controlling for firm and time fixed effects, as well as CPI inflation on the day of the firm response. Perceptions are strongly related to near-term expectations: a 1 pp increase in CPI perceptions is associated with a 0.35 pp–0.6 pp increase in near-term CPI expectations. The incomplete transmission indicates that firms do not purely extrapolate from current CPI, suggesting that firms incorporate additional information when forming expectations – for example forward-looking behaviour or belief in mean reversion towards target inflation.
In comparison to year-ahead expectations, the relationship between CPI perceptions and three-year expectations is weaker quantitatively across the sample, although still statistically significant. The weaker relationship for three-year expectations indicates that long-run beliefs are more anchored to long-term priors. The relationship is not constant over time. The sensitivity of expectations increased from late 2024 onward, coinciding with a renewed rise in inflation (grey bars). The sensitivity of three-year expectations to inflation perceptions appears to have peaked in 2025 Q3
Chart 5: CPI three-year expectations depend positively on one-year expectations
Notes: Controls: firm fixed effects, time fixed effects and CPI inflation.
In Chart 5 we analyse the relationship between one-year and three-year CPI expectations. This captures the term structure of inflation expectations: firms which update their short-run beliefs update their long-run forecasts in the same direction. The magnitude of the coefficients is below one, suggesting near-term expectations only partially correlate with medium-term expectations, and that the latter are more stable. Nevertheless, we find that the relationship has strengthened recently, indicating that medium-term expectations have become more responsive to near-term expectations as well as current CPI perceptions. While the results are not necessarily causal, they provide an insight into the expectations formation process of firms across horizon.
Own-price and CPI expectations
We next examine whether CPI expectations are correlated with firms’ expected change in the prices of their own goods and services (own price growth) in the year ahead. Understanding this link is crucial: It represents the channel through which aggregate inflation expectations translate into individual pricing decisions – ultimately influencing inflation dynamics. The link between own-price and CPI inflation expectations may be two-sided: firms may look at CPI when planning their own prices as well as extrapolate from their own-price growth to forecast inflation.
Chart 6: One-year CPI expectations matter for own-price expectations
Notes: Controls: firm fixed effects, time fixed effects and own-price growth.
As shown in Chart 6, we find a positive, significant and robust relationship between firms’ expected own-price growth and one-year CPI expectations, controlling for realised own-price growth experienced by the firm. This result suggests that near-term CPI expectations influence pricing intentions rather than merely reflecting backward-looking extrapolation. As expected, a coefficient lower than one indicates that while CPI is important for own-prices, firms recognise that their relative price can deviate from aggregate inflation based on sector-specific demand, competition and cost structure. The link with three-year CPI expectations is weaker and insignificant, consistent with longer term CPI serving more as an anchor than an input into pricing. To ensure that the results are not simply driven by common shocks, we re‑estimate the specifications using sector‑by‑time fixed effects and find that the positive links hold.
Chart 7: More productive firms react more to macro signals
Notes: Controls: firm fixed effects, time fixed effects and own-price growth. Low productivity corresponds to 0%–25% quantile and medium/high productivity corresponds to 25%–100% quantiles.
Finally, we investigate whether the link between CPI expectations and own-price expectations differs across firms, using labour productivity as the key characteristic. Chart 7 shows that for more productive firms, the relationship between one-year CPI expectations and expected own-price growth is almost twice as strong as for the bottom quartile of the distribution. Furthermore, more productive firms tend to have lower own-price growth expectations on average, consistent with productive firms facing greater competitive pressure, leading to more moderate pricing plans.
Conclusion
Firms’ perceptions and expectations play a central role in shaping inflation dynamics. UK firms have generally perceived inflation accurately, but near-term expectations adjusted more slowly during the 2023 disinflation. CPI expectations also have become more sensitive to current conditions; the increased sensitivity of three-year CPI expectations with respect to CPI perceptions and one-year expectations suggests that in our sample, expectations are less anchored than in previous years. Short-term CPI expectations matter for firms’ own-pricing. More productive firms make smaller errors, respond more to macroeconomic signals, and align CPI and own-price expectations more tightly. Our results highlight that understanding how firms form expectations across horizons, and how these expectations inform pricing, is crucial for assessing inflation persistence and the transmission of monetary policy.
Federico Pessina is a PhD student in Economics at UCL, and a PhD Intern in the Bank’s Structural Economics Division, Maren Froemel and Ivan Yotzov also work in the Bank’s Structural Economics Division.
If you want to get in touch, please email us at bankunderground@bankofengland.co.uk or leave a comment below.
Comments will only appear once approved by a moderator, and are only published where a full name is supplied. Bank Underground is a blog for Bank of England staff to share views that challenge – or support – prevailing policy orthodoxies. The views expressed here are those of the authors, and are not necessarily those of the Bank of England, or its policy committees.