The household saving rate revisited: recent dynamics and underlying drivers

Prepared by Maria Dimou, Marco Flaccadoro and Johannes Gareis

Published as part of the ECB Economic Bulletin, Issue 8/2025.

After falling back from its pandemic-related peak, the household saving rate rose again from mid-2022 to mid-2024 and has since remained broadly stable at an elevated level (Chart A). The seasonally adjusted household saving rate, as reported in Eurostat’s quarterly sector accounts (QSA), averaged around 13% between 1999 and 2019. After surging during the pandemic, in the second quarter of 2022 it returned to levels close to the historical average, before subsequently starting to rise again, reaching 15.4% by mid-2024. Since then, it has remained broadly stable at that elevated level. This box provides updated evidence on developments in the household saving rate and its recent drivers.^[1]

Chart A

Household saving rate

(percentages of gross disposable income)

Since mid-2024, households’ real disposable income and consumption have been growing at broadly similar rates, thereby stabilising the saving rate at a higher level than before the pandemic. According to a statistical decomposition, rising real income – particularly labour income – supported the saving rate prior to the pandemic, while higher real consumption exerted an offsetting effect (Chart B). Compared with the pre-pandemic period, income growth accelerated markedly from mid-2022 to mid-2024, driven mainly by stronger contributions from non-labour income – including self-employment income, net interest income, dividends and rents – and by net fiscal income related to fiscal measures introduced in response to the energy price shock, including untargeted support. These factors likely boosted the saving rate, as non-labour income accrues disproportionately to higher-income households who have a greater propensity to save.^[2] Since mid-2024 the saving rate has remained broadly stable as real income and consumption dynamics have normalised, with stronger growth in labour income – reflecting real wage catch-up and sustained employment – offsetting declining non-labour income and the gradual withdrawal of fiscal support.

Chart B

Contribution of income and consumption growth to changes in the household saving rate

(quarterly percentage changes and percentage point contributions; averages)

Empirical estimates indicate that the saving rate has remained elevated over the last year, as the negative contributions of declining real interest rates and improving real net wealth positions have not fully counterbalanced the support from strongly rising real labour income. An empirical model for real household consumption shows that, prior to the pandemic, higher real income – particularly labour income – raised the saving rate, as consumption did not adjust one-to-one with income, whereas increases in real net wealth lowered it by reducing households’ incentives to save and, thus, encouraging higher consumption (Chart C). The rise in the saving rate between mid-2022 and mid-2024 reflected the rapidly increasing labour income, while stronger-than-usual growth in other income (i.e. the sum of non-labour and fiscal income) also made a significant positive contribution.^[3] Lower real net wealth following the surge in inflation and higher real interest rates brought about by the monetary policy tightening provided additional upward contributions, while the legacy of the pandemic weighed on the saving rate as consumption normalised. Since mid-2024, real labour income growth has increased once more, providing a strong upward push to the saving rate. This has been offset, however, by rising real net wealth – whose contribution to the saving rate returned to its historical average – declining real interest rates and a reversal of earlier increases in other income.

Chart C

Contributions to the change in the household saving rate: a model-based decomposition

(quarterly percentage changes and percentage point contributions; averages)

Households’ uncertainty about their own financial situation also appears to play an important role in saving decisions. Household-level information from the Consumer Expectations Survey (CES) is used to shed light on factors not captured by standard macroeconomic determinants, as indicated by the positive unexplained component in the model-based decomposition since mid-2022 (Chart C). In particular, the analysis focuses on policy-related and individual uncertainty, as reflected in Ricardian and precautionary saving motives.^[4] A new question fielded in the November 2025 CES questionnaire reveals that precautionary and Ricardian motives are each relevant for around 50% of respondents in their saving decisions, with 25-30% also indicating one or the other as the most important reason to save (Chart D, panel a).^[5] A closer look at cross-sectional differences across participants shows that the importance of these two motives depends mainly on the economic constraints faced by respondents and the degree of uncertainty about their financial situation, while differences across income and age seem to play a more muted role (Chart D, panel b). Households with no self-reported liquidity constraints are more likely to attach significantly higher importance to both motives, consistent with their stronger capacity to plan and save (“unconstrained”). At the same time, respondents reporting a high degree of certainty about their future financial situation (“certain”) attach significantly less importance to the two motives than households facing greater uncertainty. The similarities in the determinants of the two motives would suggest that they are close conceptually, with respondents tending to perceive the actions of the government as an additional source of income uncertainty. This is consistent with previous analysis that has highlighted the importance of geopolitical and policy uncertainty for consumer spending.^[6]

Chart D

Prevalence of the precautionary and Ricardian motives among respondents and its determinants

a) Prevalence of precautionary and Ricardian motives across survey respondents

(relevance score (0-100) and percentages of respondents)

b) Determinants of prevalence of precautionary and Ricardian motives

(change in mean relevance score, percentage points)

Overall, model and survey-based evidence suggests that both conjunctural and behavioural factors have contributed to the saving rate remaining elevated but broadly stable. While income and wealth dynamics have largely returned to pre-pandemic patterns, heightened uncertainty and precautionary motives have likely continued to exert upward pressure on savings.

References

Andersson, M., Bobasu, A. and De Santis, R.A. (2024), “What are the economic signals from uncertainty measures?”, Economic Bulletin, Issue 8, ECB.

Bankowski, K., Bouabdallah, O., Checherita-Westphal, C., Freier, M., Jacquinot, P. and Muggenthaler, P. (2023), “Fiscal policy and high inflation”, Economic Bulletin, Issue 2, ECB.

Barro, R.J. (1974), “Are government bonds net wealth?”, Journal of Political Economy, Vol. 82, No 6, pp. 1095-1117.

Bobasu, A., Gareis, J. and Stoevsky, G. (2024), “What explains the high household saving rate in the euro area?”, Economic Bulletin, Issue 8, ECB.

Carroll, C.D. (1997), “Buffer-stock saving and the life cycle/permanent income hypothesis”, The Quarterly Journal of Economics, Vol. 112, No 1, pp. 1-55.

Ceci, D. and Flaccadoro, M. (2026), “The recent weakness in household consumption: evidence from the euro area and Italy”, Occasional Papers, Bank of Italy, forthcoming.