Cash is alive… and somewhat young? Decoupling age, period and cohort from euro cash use

Prepared by Rebecca Clipal and Alejandro Zamora-Pérez

1 Introduction

A central puzzle motivating modern payments research is the divergence between decades of predictions about a cashless society and the empirical reality of the enduring role cash plays. This persistence of cash is paradigmatic in the case of the euro. The value of euro cash circulation reached €1.59 trillion by the end of 2024, representing around 10% of euro area GDP (a similar share to that of a decade earlier). ^[1] While the transactional use of cash – estimated to account for roughly 20% of this circulation (Zamora-Pérez, 2021) – has clearly diminished with ongoing digitalisation, this trend has been slower than predicted and the demise of cash has not materialised. To the contrary, evidence shows that cash was used in about 50% of physical point-of-sale transactions in 2024, remaining people’s most frequently used payment method with the widest merchant acceptance (European Central Bank, 2024a; European Central Bank, 2024b). This resilience contradicts decades of predictions about the death of cash (Shy, 2023), which suggest that other factors beyond simple transactional efficiency influence overall demand for physical currency.

What explains this rarely anticipated persistence of cash and what are the future prospects of cash? Early predictions appear to have underestimated several overlapping factors: (1) the imperfect substitutability of cash with electronic payment methods (Alvarez, et al., 2022, Brown, et al., 2022), (2) the stickiness of payment habits, owing to the costs associated with adopting new means of payment (Van der Cruijsen et al., 2016; Nocciola and Zamora-Pérez, 2024), and (3) the continued relevance of cash for varied demographic groups and payment scenarios (Bagnall, et al., 2016), including for users with full access to digital payments in advanced economies (Zamora-Pérez et al., 2024) and groups at risk (Van der Cruijsen and Reijerink, 2024).^[2] Forecasts also focused solely on the payment function of cash and underappreciated other roles of cash, such as its use as a tangible store of value, which has been increasing. ^[3] Beyond these explanations for the overall resilience of cash, another key factor is the differentiated use of varied functions of cash across age groups. This remains a less explored area and offers an opportunity for deeper insight, particularly by systematically analysing interactions between ageing processes and broad societal changes, like digitalisation.

This article contributes to the discussion by examining the current role played by cash across three of its functions, decoupling period, age and cohort effects for each of them. This article focuses on cash in everyday transactions, its function as a store of value and the importance individuals place on retaining cash as an option (sometimes referred to as latent cash demand). These are key indicators of the vitality of cash in the euro area. Using data from the Eurosystem SPACE survey (2019, 2022 and 2024), we decouple period effects (such as general digitalisation trends or singular events like the pandemic) from age-specific patterns and the life-cycle dynamics of distinct birth cohorts, with the aim of understanding the underlying drivers of the persistent – albeit evolving – relevance of cash. Our findings show clear period effects and stable age patterns over the last five years: younger individuals engage with cash across all functions, particularly holding it for precautionary reasons, whereas older groups use it for transactions more frequently than other age groups. The perceived importance of cash has risen for all age groups.

2 A descriptive overview of cash functions by age and period (2019-24)

A first look at the descriptive data suggests that cash use is not monolithic – the ways cash is used vary depending on temporal trends and people’s age. We draw on the 2019, 2022 and 2024 editions of the SPACE survey.^[4] Our analysis encompasses approximately 110,000 adults across 15 euro area countries and – via representative payment diaries and questionnaires – incorporates rich cross-sectional data on payment choices, cash holdings and attitudes.^[5]

The share of cash in everyday payments (including physical and online transactions) decreased across all age groups between 2019 and 2024. Chart 1 reveals a uniform decline between 2019 and 2022 which was particularly pronounced, likely reflecting shifts driven by the pandemic. There was more variation from 2022 to 2024. Younger age groups, for example 18–22 and 23–27 years, saw a steady decline in cash use over the three years, whereas for middle and older age groups, the trend was more moderate, with a less pronounced drop in both volume and value between 2022 and 2024. Throughout all periods, older age groups consistently maintained a higher share of cash transactions compared with younger cohorts – these increasingly favoured online cashless payments. For most younger and middle-aged groups, the combined share of online and physical cashless transactions surpassed that of cash by 2024, with online transactions rivalling cash for some young demographics, e.g. 23-32 years. These findings point to persistent age-related preferences, as well as strong period effects driving digitalisation.

Chart 1

Payment methods in everyday transactions by age group and year

(age and period; percentage)

Beyond payments, cash is increasingly used as a store of value and its reported importance among consumers is also on a distinct upward trend. Chart 2 presents two heatmaps where darker shading signifies a higher proportion of individuals within a given age-period cell. Panel a) displays the shares of individuals with precautionary cash reserves held at home. It shows a visible increase across all age groups in 2022, suggesting a strong period effect likely linked to the pandemic. Additionally, younger individuals (18-22 and 23-27 years) consistently report higher levels of precautionary cash holdings across all three waves, which indicates a distinct age effect. Panel b) shows a clear increase in the perceived importance of cash as a payment option from 2019-24 across age groups, indicating a period effect. However, the youngest and oldest cohorts display slightly different patterns, suggesting a more nuanced age effect. By 2024, cash is perceived as important by a wide majority of individuals across all age groups.

Chart 2

Reported use of cash as a store of value and the perceived importance of cash by age and survey

a) Use of cash as a store of value	b) Perceived importance of cash
(right-hand scale: percentage of individuals holding cash at home; left-hand scale: age in years; x-axis: year)	(right-hand scale: percentage of individuals considering cash important; left-hand scale: age in years; x-axis: year)

3 Period, age and cohort effects decoupled

To disentangle the influences observed above, this article applies an age-period-cohort analytical framework to data from three editions of the SPACE survey. While the surveys do not track individual consumers over time, the methodology allows for the statistical decomposition of observed patterns into three key components. These components are: (1) age effects that capture variations linked to individuals’ life stages, (2) period effects that reflect shocks or trends affecting all individuals at a given point in time, such as the pandemic or ongoing digitalisation, and (3) cohort effects that refer to persistent differences between people born around the same time who may develop distinct behavioural patterns shaped by shared experiences.^[6]

The age-period-cohort decomposition, though not widely applied to the analysis of retail payments, is crucial for understanding that an apparent uniform shift in cash use can mask diverse underlying mechanisms. This framework first calculates an overall average, i.e. a grand mean, of cash use across all respondents and survey years. It then isolates the effects of age and period by measuring how each deviates from this baseline. Importantly, it also detects cohort-specific behaviours by examining whether a given cohort consistently uses cash more or less than would have been expected based on their age or the period alone.

This section focuses on the main period and age effects influencing cash use and attitudes, before examining cohort-specific dynamics. We assess three key indicators: transactional cash demand, cash held as a store of value and the subjective importance attributed to having cash as a payment option. Each of the three indicators is presented in a two-panel chart, displaying both period and age effects.^[7] In addition to the baseline effects, we also test the influence of a set of demographic, financial, technological, payment-related and persistent country-related factors and variables to understand what might be driving the patterns.^[8]

3.1 The baseline: period and age effects

Cash use for everyday transactions

The share of cash in day-to-day payments (physical and online) has fallen over time (from 2019-24), yet it follows a distinct age-related pattern that dips in the mid-twenties and climbs in later years.^[9] Panel a) in Chart 3 shows a strong period effect with respect to the grand mean depicted by the horizontal zero line, and it is statistically significant. The black error bar for each year does not intersect the zero line. The solid blue line is well above zero in 2019 and far below zero in 2024. This signals a large, statistically systemic fall in cash use over time, which we normally refer to as overall digitalisation in payments. Panel b) shows that the youngest age groups show average levels of cash use for transactions. This falls for individuals from their late twenties to late forties (indicating lower than average cash use) and rises clearly above the mean for those older than 60.

Chart 3

Estimated period and age effects on transactional cash share in everyday payments

a) Period effects	b) Age effects
(percentage-point deviation from overall mean cash share)	(percentage-point deviation from overall mean cash share)

There is a sharp observed decline across periods significantly driven by the surge in online transactions, while the V-shaped “tick” age pattern is influenced by life-cycle factors. The period-driven decline in transactional cash share in panel a) of Chart 3 was most acute between 2019 and 2022, largely reflecting the rapid adoption of online and contactless payments accelerated by the pandemic. A further, though less steep, reduction in cash share occurred by 2024, even as online payments also began to replace physical cashless methods (see Chart 1). Regarding the age dimension, the tick pattern suggests typical life-cycle transitions: younger individuals exhibit average cash reliance, individuals in their prime working years make greater use of digital and physical cashless methods, and people of older age discernibly use cash more.^[10]

Controlling for employment status (the dotted lines in Chart 3) noticeably flattens both the period and age effect curves, indicating its relevance for understanding age and group effects. As regards periods, the findings suggest that employment – encompassing factors like income regularity, workplace payment norms and the transition into retirement – mediates cash usage. A decrease from 2019 in aggregate unemployment levels across periods likely changed the composition and behaviours of employed payers, which contributes to the overall period trend in cash use.^[11] With regard to age, younger individuals entering employment earlier might use cash less than their student peers, while employed older individuals might use cash more than retirees in the same age bracket.

Using cash as a store of value

Immediately after the pandemic, more people kept some cash at home, and this habit is surprisingly most prevalent among the youngest age groups.^[12] Panel a) of Chart 4 shows a one-off period effect: the baseline jumps well above the zero reference line in 2022 and drops below it in 2024, with error bars clear of the line at both dates. Panel b) depicts the age effects: the baseline model (solid line) shows that the youngest age groups (18-37 years) have a strikingly high and statistically significant propensity to store cash at home. This tendency diminishes rapidly and significantly through early adulthood, reaching its lowest point for those in their late 60s to late 70s, before trending upwards for the oldest cohort whose behaviour is indistinguishable from the mean.

High uncertainty (between 2019 and 2022) and higher opportunity cost (between 2022 and 2024) mostly explain the period shifts in store of value, while youth-specific factors explain the age pattern. The pronounced peak in cash hoarding in 2022 likely reflects a flight to perceived safety and liquidity during the acute phase of the pandemic, as noted in previous literature (Tamele, et al., 2021). Its subsequent decline by 2024 may be attributable to a normalisation of perceived risk and (more critically) the rising opportunity cost of holding non-interest-bearing cash as the ECB began increasing interest rates from July 2022 onwards.^[13] The persistent and significant cash hoarding among the youngest adults is particularly noteworthy. This could stem from several interlinked factors: lower engagement with formal financial institutions, receipt of part of income, gifts or allowances in cash rather than from wage accounts, and strong parental reliance, e.g. because of the high average age at which people leave their parental homes in Europe, and in some euro area countries specifically.^[14]

Employment status accentuates the pandemic spike but hardly moves the age curve, and no other covariate we tested changes the picture. Interestingly, controlling for employment status – the dotted line in Chart 4, panel a) – significantly exacerbates the 2022 peak in the period effect. This suggests that those not in employment may have had an even stronger precautionary motivation to hoard cash. Our analysis found that no other demographic, financial, technological, or payment-related factors significantly altered the fundamental age pattern observed in panel b). This robustness points to the age effect being a deeply entrenched behavioural characteristic, reflecting distinct life-cycle stages and attitudes towards financial management, rather than being easily explained by more conventional socioeconomic indicators.

The importance of being able to pay with cash

The perceived importance of cash as a payment option has risen in every edition of the SPACE survey, and this rise is common across age groups.^[15] Across the observed periods, individuals increasingly report that having the option to pay with cash is important. Panel a) of Chart 5 depicts this period effect: the solid line starts significantly below the mean in 2019, crosses the zero line to become statistically indistinguishable from the mean in 2022 and rises to a statistically significant positive deviation by 2024. However, this indicator shows no age effect, suggesting that the sentiment is broadly shared across age groups with no statistically significant age-specific deviations – see panel b). Only the 33-37 age group displays a statistically significant deviation from the grand mean. This suggests a growing perception that cash is important; it is a widespread phenomenon, not strongly differentiated by age.

Chart 5

Estimated period and age effects on the perceived importance of cash

a) Period effects	b) Age effects
(percentage-point deviation from overall mean cash share; year)	(percentage-point deviation from overall mean cash share; age in years)

Several hypotheses could also account for the broad rise in the perceived importance of cash. Increased societal awareness of vulnerabilities of digital systems and cyber threats might bolster the appeal of cash attributes.^[16] These vulnerabilities and threats include misuse of data shared with third parties, privacy breaches related to digital transaction traces, and the impact of online fraud. Furthermore, it is possible that the impact of recent crises (such as pandemic-related disruptions, Russia’s invasion of Ukraine, power outages or other non-systemic failures affecting digital payments) could be reinforcing the view that cash is a resilient and relevant backup (Faella and Zamora-Pérez, 2025). All these are potential explanations for the observed period trend, as our direct controls do not capture such specific sentiments.

Only two observable factors do notably change the baseline effects. Controlling for employment status – the dotted line in panel a) – significantly exacerbates the upward period trend. This suggests that, conditional on employment, the perceived importance of cash grew even more strongly. When controlling for internet use – the dotted line in panel b) – the previously flat age pattern transforms. The line now slopes significantly downwards from younger to older age groups, indicating that, among individuals with similar levels of internet use, younger people report cash as being of significantly higher importance than is reported by older internet users. This suggests that for people who use the internet heavily (a group that tends to be disproportionately younger) the stated importance of cash might be a conscious counterpoint or a retained value despite their digital immersion, whereas for older internet users, their digital engagement might more fully supplant the perceived need for a cash option. No other demographic, financial or payment covariate shifts the picture. This highlights that the period effect is broad-based and the age effect genuinely small.^[17]

3.2 Cohort effects: does any specific generation deviate?

In addition to period and age effects, we also assess if specific generational cohorts (people born around the same time) display unique behavioural or attitudinal trajectories. This involves assessing two key considerations: (1) whether a specific birth cohort consistently behaves differently to other cohorts of the same age and in the same period (inter-cohort effect) and (2) whether their behaviour has changed as they age (intra-cohort effect)?^[18] Identifying these two cohort effects matters, as it ensures that patterns primarily driven by generational differences are not misattributed solely to age or period. This ultimately enables more accurate decomposition of the observed developments.^[19]

Our model-based estimates identify, at best, weak cohort-specific deviations for cash transaction and store-of-value behaviour, which further emphasises the relevance of period and age effects. Most of the changes in cash transactions are explained by age and the period, as cohort effect patterns were not globally significant. For precautionary cash holdings at home, weak cohort differences emerge, mainly driven by the youngest generational group – those born in the early 2000s – who stored less cash than expected for their age (inter-cohort effect). Visible intra-cohort effects are also observed. For example, the share of people born in the mid-to-late 1990s who have cash holdings at home has significantly decreased over five years relative to the age-period baseline. This suggests unobserved factors are uniquely affecting the life-cycle dynamic of this specific cohort.

The perceived importance of cash exhibits the clearest cohort effects. Our previous analysis showed that the importance people place on having cash as an option has risen over time (a period effect), as most age groups appear to share this increasing sentiment somewhat uniformly (with no overarching age effect). However, when we look at generational cohorts, more complicated patterns are observed, with no simple young versus old pattern. This suggests that the year a person is born significantly shapes how they perceive cash, possibly because stated preferences are more malleable than ingrained payment or saving habits, which often exhibit more stable age-related patterns. For example, the oldest cohort (born in the late 1940s) and the cohort born in the late 1990s consistently value cash more highly than average. In contrast, some cohorts (like those born in the early 1950s and early 2000s) place less importance on cash. Beyond these baseline generational stances, we also observe attitudinal shifts within cohorts as they age.^[20] Hence, the rising importance of cash as an option appears less tied to age effects and more reflective of the shifting attitudes of different birth cohorts.

3.3 A visual summary of the results

Combining age, period and cohort effects, we can visually track how cash behaviours and attitudes of specific birth cohorts evolve as they age across editions of the survey. Chart 6 visualises these dynamics, showing model-predicted probabilities for our three indicators.^[21] Each curve represents a different birth cohort (the youngest, born in the late 1990s, in yellow, and the oldest, born in the 1940s, in blue) as they age through the five years we have analysed. In this chart, the period, age and cohort effects that we disentangled above are combined, so only the overall summary effect can be seen clearly.

Chart 6

How cash habits change as we age: a model-based view across birth cohorts

(predicted probabilities)

Cash transactions tend to be higher for older generations (except for the youngest) and all birth cohorts decreased their cash use over time at different speeds. Chart 6, panel a) shows an overall upward trend, indicating that older generations tended to maintain a higher share of cash transactions at any given age compared with younger generations. The youngest birth cohort is an exception to this pattern, as it is predicted to use cash slightly more often than the second youngest cohort, partly reflecting the tick age pattern discussed above. Simultaneously, a general decline in the cohort lines across the three periods points to a decreasing overall use of cash over time indicating the overall trend towards digitalisation. The predicted cash shares tend to converge, showing how digitalisation affects generations differently.

Younger cohorts stored cash more often than older ones. Panel b) illustrates the probability of individuals storing cash. It reveals that younger generations start more likely to store cash, but this declines sharply with age. The common bell shape across cohort lines also points to increased cash hoarding during the pandemic, with more varied cohort trajectories suggesting some generation-specific responses.^[22]

Generally, all cohorts increasingly view cash as important, albeit at varying celerity. The perceived importance of cash, as seen in panel c), shows consistently high valuation across all ages (ranging from above 50% to around 65%) and generally increasing over time. Most cohort lines also trend upwards, indicating cash increasingly being seen as important, especially for intermediate cohorts.

4 Cash use in a digital-analogue hybrid future

The differentiated but enduring cash demand highlighted by this article suggests that payment policy frameworks need to remain cash-compatible in the future. Our analysis of granular data, across transactional and store-of-value functions and perceived importance, reveals that cash fulfils persistent roles in a different manner for different age groups. Given this assessment, it is likely that the demand for different functions of cash will continue to evolve across demographics in the coming decade.

Our findings on age effects – for example, younger groups exhibiting notable engagement across the three analysed dimensions – can help inform policy by offering a better understanding of the ways people use and think about cash. The age patterns for transactions, the tendency for younger adults to store cash at home, and the shared perception (potentially heightened for online younger people) that it is important to have cash as an option, challenge narratives of a unidirectional shift towards a cashless society. Age is clearly associated with variations in cash use. As younger people enter the labour market, establish independent living and manage early-career finances, cash appears to serve important functions.^[23] Middle-aged groups display distinctive transactional patterns, in particular a stronger decrease in cash transactions than is the case for other cohorts during the five years analysed. Older age groups consistently demonstrate higher transactional cash use, lower store-of-value behaviour and similar views as the other cohorts on the importance of cash, which tend to remain robust or increase.

While our findings suggest the role of cash is evolving, future demand for cash is difficult to forecast because of interacting demographic and economic trends, uncertain one-off events and the regulatory stance toward cash. Observed age effects suggest that future generations may continue to rely on cash as they do today, which would mean cash is more likely to continue being used. However, euro area population ageing patterns mean that the behaviours of middle-aged and especially older cohorts will carry increasing weight in aggregate cash demand (Eurostat, 2025).^[24] This may reveal a trend that counteracts the ongoing, mostly universal, shift towards digitalisation, based purely on demographic considerations and findings linked to stable age effects of older individuals. However, unforeseen events, such as heightened geopolitical uncertainty, concerns about cybersecurity or data privacy associated with digital payments, or even significant disruptions to digital infrastructures, could substantially shift public perceptions and cash use during specific periods (Faella and Zamora-Pérez, 2025). In this context, the regulatory stance regarding cash (such as imposing payment limits) and the positioning of the industry (such as closing or opening cash access points) will also play an important role. Policies focused on ensuring cash remains universally accepted, and that support solutions which preserve ease of use and access it, will significantly influence the ways cash is used in the future.

Given these dynamics, a key challenge for both public and private sectors is to ensure the continued viability and accessibility of the cash infrastructure, even when transaction volumes potentially decrease. Declining cash transaction volumes, if they were to continue unabated, could pose challenges to the cost-effectiveness of maintaining the existing infrastructure and day-to-day operations when it comes to cash. Yet, the findings presented in this article underscore the continuing importance of cash in ensuring payment choice across age groups, in contributing to financial inclusion and in serving as a resilient failsafe means of payment. This is reflected in the Eurosystem cash strategy (European Central Bank, 2025), which acknowledges the enduring appeal of cash to different population segments for specific uses in a range of circumstances.^[25] As discussions about promoting digitalisation and innovation continue, the evidence presented here suggests that such initiatives should be framed to accommodate the likely future persistence of physical cash use. This would help acknowledge the unique benefits of cash while embracing the full range of innovation in payments. It is an approach that would foster a robust, efficient payment ecosystem that serves everyone at all times and in all situations.

Strategic policy approaches are therefore warranted to ensure cash remains a viable, accessible and affordable option, while also seeking to avoid unnecessarily increasing the costs associated with handling cash. An instructive example is Sweden, where an accelerated cash decline – partly attributed to policies such as strict cash changeovers and measures to combat tax evasion (Sveriges Riksbank, 2020) – led to public concerns and parliamentary debates. These concerns ultimately prompted actions aimed at legislating to preserve access to, and ensure the acceptance of, cash (Sveriges Riksdag, 2023; Sveriges Riksbank, 2025). This speaks in favour of: (1) monitoring the cost and geographic reach of the cash-circulation cycle, fostering solutions to make access remains feasible even if volumes drift lower (Zamora-Pérez, 2022); (2) reviewing crisis preparedness strategies, incorporating cash as an analogue backup tool and developing business-continuity plans in case digital channels are disrupted; (3) understanding the adverse effects of policies that restrict cash use; and (4) clarifying how legal-tender rules apply, i.e. whether merchants or public services are allowed to adopt card-only models.^[26] Findings on the use of cash by younger people likewise suggest that financial inclusion and education initiatives might still benefit from covering physical money alongside digital tools. Finally, the fact that households continue to hold cash as a safe asset, despite not bearing interest, is worth keeping in mind for monetary and payment policy discussions (Faella and Zamora-Pérez, 2025).

5 Conclusion

The evidence from three editions of the SPACE survey and our age-period-cohort framework offer new findings on the evolving (transactional and store-of-value) use of cash and its perceived importance. Overall trends, like digitalisation or an increased perception that cash is important, and one-off events, like the pandemic, can markedly alter people’s behaviour. Furthermore, age effects and life-cycle dynamics of specific generations are critical for understanding evolving cash use and perceptions. The temporary surge in cash hoarding during the pandemic illustrates how long-standing preferences or perceptions can manifest themselves strongly under certain conditions.

An important point for understanding present and future cash demand is that cash serves multiple functions and its utility varies significantly across age groups. While ongoing payment digitalisation is reducing cash transactions, cash concurrently exhibits sustained and period-driven fluctuations in its rising perceived importance and its role as a store of value. Predictions from 20 years ago regarding the swift or complete replacement of cash have not yet materialised, partly because these did not consider the varied ways cash would continue to be useful to people. In this article, we show that all these indicators of cash preferences display unique age patterns that can contribute to a better understanding of the persistence in cash use across different functions.

For example, younger people’s high engagement with cash suggests that new generations may continue to find cash useful as they enter the job market, although the effects of an ageing population should also be considered. Younger people notably use cash across its three functions – exhibiting average transactional use, playing a leading role in precautionary cash storage and expressing a strong view that cash is important. This indicates that cash is considered useful during the early stages of financial independence. Simultaneously, it may be the case that the higher volume of cash transactions observed among older generations in ageing euro area populations is a trend that pushes back against the speed of digitalisation. Even if every cohort were to retire with a lower intensity of cash use than its predecessor (period effect), demographic considerations (age effects) may imply a slowing – not accelerating – aggregate decline. All this sheds further doubt on the maximalist position that expects a steep decline in the use of cash in the coming years.

A prudent payment policy stance must understand the utility that cash provides to society and plan for a long phase of coexistence, in which banknotes and coins circulate alongside electronic, physical and online payment methods. Alongside the undeniable benefits of digitalisation and its potential to enhance financial inclusion, policies or industry strategies that unduly restrict access to cash or make its use more expensive, risk disadvantaging parts of the population and weakening systemic resilience; conversely, ignoring digital progress would miss efficiency gains. Balanced strategies that safeguard essential cash infrastructure and encourage competitive digital options, while remaining attentive to demographic diversity, will best serve the euro area as payment habits evolve.

References

Alvarez, F. and Argente, D. (2022), “On the Effects of the Availability of Means of Payments: The Case of Uber”, The Quarterly Journal of Economics, Vol. 137, No 3, August, pp. 1737-1789.

Alvarez, F. et al. (2022), “Cash: A blessing or a curse?”, Journal of Monetary Economics, Vol. 125, January, pp. 85-128.

Bagnall, J. et al. (2016), “Consumer Cash Usage: A Cross-Country Comparison with Payment Diary Survey Data”, International Journal of Central Banking, Vol.12, No 4, December, pp. 1-61.

Brown, M. et al. (2022), “The Convenience of Electronic Payments and Consumer Cash Demand”, Journal of Monetary Economics, Vol. 130, September, pp. 86-102.

Chodorow-Reich, G. et al. (2020), “Cash and the Economy: Evidence from India’s Demonetization”, The Quarterly Journal of Economics, Vol. 135, No 1, February, pp. 57-103.

European Central Bank (2024a), Study on the payment attitudes of consumers in the euro area (SPACE), December.

European Central Bank (2024b), Use of cash by companies in the euro area, September.

European Central Bank (2025), The Eurosystem cash strategy.

European Union Agency for Cybersecurity (ENISA, 2024), “2024 Report on the State of the Cybersecurity in the Union”, ENISA Reports, ENISA, December.

Eurostat (2024), Age of leaving the parental home steady in 2023, 30 September.

Eurostat (2025), “Population structure and ageing”, Statistics Explained, Eurostat.

Faella, F. and Zamora-Pérez, A. (2025), “Keep calm and carry cash: Lessons on the unique role of physical currency across four crises”, Economic Bulletin, Issue 6, ECB, forthcoming.

Lagos, R. and Zhang, S. (2022), “The Limits of ONETARY ECONOMICS: On Money as a Constraint on Market Power”, Econometrica, Vol. 90, No 3, May, pp. 1177‑1204.

Luo, L. and Hodges, J.S. (2022), “The age-period-cohort-interaction model for describing and investigating inter-cohort deviations and intra-cohort life-course dynamics”, Sociological Methods & Research, Vol. 51, No 3, August, pp. 1164-1210.

Nocciola, L. and Zamora-Pérez, A. (2024), “Transactional demand for central bank digital currency”, Working Paper Series, No 2926, ECB, April.

Ryder, N.B. (1985) “The Cohort as a Concept in the Study of Social Change”, in Mason, W.M. and Fienberg, S.E. (eds), Cohort Analysis in Social Research, Springer, New York, NY, pp. 9-44.

Shy, O. (2023), “Cash is alive: How economists explain holding and use of cash”, Journal of Economic Literature, Vol. 61, No 4, December, pp. 1465-1520.

Sveriges Riksbank (2020), “Why are people in Sweden no longer using cash?”, Payments in Sweden 2020, 29 October.

Sveriges Riksbank (2025), “Introduce obligation to accept cash and strengthen banks’ responsibility for cash”, press release, 19 May.

Sveriges Riksdag (2023), “Lag (2010:751) om betaltjänster” (only available in Swedish).

Tamele, B. et al. (2021), “Catch me (if you can): assessing the risk of SARS-CoV-2 transmission via euro cash”, Occasional Paper Series, No 259, ECB, July.

Van der Cruijsen, C., Hernandez, L. and Jonker, N. (2016), “In love with the debit card but still married to cash”, Applied Economics, Vol. 49, No 30, 15 November, pp. 2989-3004.

Van der Cruijsen, C. and Reijerink, J. (2024), “Uncovering the digital payment divide: Understanding the importance of cash for groups at risk”, The Journal of Consumer Affairs, Vol. 58, No 2, Summer, pp. 486-505.

Yang, Y. and Land, K.C. (2013), Age-Period-Cohort Analysis: New Models, Methods, and Empirical Applications, Chapman and Hall/CRC, New York.

Zamora-Pérez, A. (2021), “The paradox of banknotes: understanding the demand for cash beyond transactional use”, Economic Bulletin, Issue 2, ECB.

Zamora-Pérez, A. (2022), “Guaranteeing freedom of payment choice: access to cash in the euro area”, Economic Bulletin, Issue 5, ECB.

Zamora-Pérez, A., Marini, A. and Honkkila, J. (2024), “Is there a digital divide in payments? Understanding why cash remains important for so many”, Economic Bulletin, Issue 2, ECB.