Aging and the Timing of Intergenerational Transfers

J. IGNACIO CONDE-RUIZ

FRANCISCO GARCÍA-RODRÍGUEZ

Estudios sobre la Economía Española 2026/16

Abril 2026

J. Ignacio Conde-Ruiza† and Francisco García-Rodríguezb aUniversidad Complutense de Madrid, ICAE and FEDEA bUniversidad de Alcalá and EQUALITAS

April 2026

Abstract

This paper shows that demographic aging reshapes intergenerational wealth transmission not only by affecting the size of transfers, but—crucially—by delaying the timing at which they are received, with important implications for wealth inequality and intergenerational mobility. While existing research has emphasized the role of inheritances in wealth accumulation, we argue that delayed inheritance receipt has become a central and unequal channel through which economic advantage and disadvantage persist across generations. Using microdata from the Spanish Survey of Household Finances (2002–2022), we show that the average age at first inheritance has increased by almost twenty years over recent decades. We quantify the economic cost of delayed inheritances, finding that each additional year of delay is associated with approximately over a 2% reduction in long-run net wealth. A simple model and stratified estimates by education reveal that this timing penalty is concentrated among households with low and medium education, consistent with the presence of binding credit constraints. We further show that earlier inheritances shape key life-cycle decisions, increasing the likelihood of homeownership, entrepreneurship, and investment in other real estate. Finally, we examine early inter vivos transfers, which could in principle provide early access to inherited resources but are disproportionately received by wealthier households and generate economically meaningful benefits primarily for the highly educated. Overall, our findings indicate that demographic aging amplifies wealth inequality by weakening the equalizing role of intergenerational transfers and concentrating their economic benefits among already advantaged groups.

Keywords: Intergenerational transfers; Inheritances; Credit constraints; Wealth inequality; De mographic aging; Inter-vivos transfers.

JEL Classification: D31; D14; E21; J10; G51.

∗We thank Luis Bauluz, Olga Cantó, Raquel Carrasco, Angel de la Fuente, Vincenzo Galasso, Juan F.´ Jimeno, Juan C. Palomino, Miguel Parra, Dmitry Petrov, Juan G. Rodríguez, Pedro Salas-Rojo, Tobias Thomas and Ernesto Villanueva for helpful comments and suggestions. We also thank seminar and conference participants at Madrid Inequality Network, University of Valladolid and European Public Choice Society for valuable feedback. José Ignacio Conde-Ruiz acknowledges the support of the Research Project of the Ministry of Science and Innovation, PID2023-148090NB-I00.

†Corresponding Author: José Ignacio Conde-Ruiz, FEDEA, Calle Jorge Juan 46, 28001, Spain; E-mail: nacho.conderuiz@gmail.com

1 Introduction

Inheritances play a central role in wealth accumulation, shaping both the intergenerational distribution of resources and patterns of intergenerational mobility. By enabling recipients to acquire assets, repay debts, or invest in new opportunities, intergenerational transfers constitute a key mechanism through which wealth is transmitted across generations and, ultimately, concentrated. As a result, inheritances can either mitigate or reinforce existing inequalities, depending on who receives them, how much is transferred, and how these resources are deployed over the life cycle. The economic impact of intergenerational transfers therefore depends not only on their size, but also on the timing of their receipt. As life expectancy rises and wealth transfers are increasingly delayed, the timing of inheritances has become as decisive as their magnitude for long-run wealth dynamics, social mobility, and the concentration of assets.

A large literature has examined the implications of inheritances from three complemen tary perspectives. A first strand documents how inheritances foster wealth concentration. Using administrative data, Boserup et al. (2016) and Elinder et al. (2018) show that bequests increase absolute inequality, primarily by bolstering the wealth of households already at the top of the distribution, while Piketty (2011); Piketty and Zucman (2015) trace how inheri tances have historically contributed to persistent asset concentration in advanced economies. A second strand examines how households adjust labor supply, saving, and portfolio deci sions in response to intergenerational transfers and wealth shocks. This literature includes both micro-level evidence on behavioral responses to specific windfall events (e.g., Elinder et al., 2012; Garbinti and Georges-Kot, 2019; Spiteri and von Brockdor!, 2023) and broader life-cycle analyses of how transfers interact with saving behavior and wealth accumulation over time (e.g., Modigliani, 1988; Gale and Scholz, 1994; Black et al., 2025). Large transfers may relax liquidity constraints, prompt earlier retirement, alter saving patterns, and shift portfolio composition toward riskier assets while reducing outstanding debt (Spiteri and von Brockdor!, 2023). A third strand emphasizes the role of credit constraints. Family transfers finance a substantial share of down-payments for first-time homebuyers (Engelhardt and Mayer, 1998), unexpected bequests reduce household leverage (Druedahl and Martinello, 2022), and inheritances facilitate entry into financial markets among liquidity-constrained individuals (Andersen and Nielsen, 2011). Related evidence highlights the close relationship between human capital and credit access, documenting that individuals with higher educa tional attainment tend to face fewer borrowing constraints (Lochner and Monge-Naranjo, 2011; Popov, 2014; Mogstad, 2017; Pfeffer, 2018).

Despite these advances, existing research has largely treated inheritances as one-o! wealth shocks, abstracting from the life-cycle stage at which transfers occur. While the literature provides extensive evidence on what inheritances do—how they affect wealth accumulation, behavior, and portfolio choices—it remains largely silent on whether these ef fects differ systematically depending on when in the life cycle inheritances are received. This omission is particularly consequential in a context of demographic aging. Rising longevity mechanically delays bequests, potentially weakening their capacity to relax credit constraints at critical stages of the life cycle, shape key economic decisions, and foster upward mobility. As a result, delayed inheritances may contribute to rising inequality not only through their unequal distribution, but also through their diminished economic effectiveness for certain groups.

Moreover, much less is known about the extent to which families can anticipate delayed inheritances through inter vivos transfers. Unlike bequests, which occur involuntarily at death, gifts reflect active donor decisions and may serve as a partial hedge against the economic costs of delayed inheritance receipt, particularly when households face binding liquidity constraints (see, e.g., Guiso and Jappelli, 2002). At the same time, the ability to transfer resources early depends on long-run family wealth and life-cycle considerations on the donor side (see, e.g., Cox, 1987; Hurd et al., 2011). This creates a second, distinct channel through which demographic aging may amplify inequality: while delayed inher itances disproportionately penalize households with fewer resources and lower access to credit, wealthier families may protect their descendants by shifting transfers forward in time. Anticipatory transfers can thus attenuate timing penalties for some households while reinforcing wealth concentration and intergenerational persistence for others.

This paper contributes to the literature by explicitly focusing on the timing of inter generational transfers and showing that demographic aging fundamentally reshapes their distributional and mobility implications. We argue that as inheritances are received later in life, their capacity to foster wealth accumulation, relax credit constraints, and shape major economic decisions is systematically weakened—especially for households with more limited access to credit. By contrast, households with greater self-financing capacity are largely insulated from these timing effects, highlighting the unequal consequences of demographic aging across socioeconomic groups.

Our main results can be summarized as follows. We show that the timing of inheritances has large and economically meaningful effects on long-run wealth accumulation: each ad ditional year of delay in inheritance receipt is associated with a reduction in net wealth of around 2%, implying sizeable losses when transfers are postponed by a decade or more. This timing penalty is highly heterogeneous and concentrated among households with low and medium education, for whom receiving an inheritance earlier in life is associated with more wealth accumulation, whereas inheritance timing plays little role for highly educated households, consistent with differences in access to credit and self-financing capacity. We also find that early inheritances shape key life-cycle decisions with long-run consequences, as households that receive inheritances earlier are significantly more likely to become home owners, invest in additional real estate, and enter entrepreneurship, with effects that are again strongest among more credit-constrained groups. Strikingly, among primary-educated households the timing of inheritance receipt predicts the acquisition of additional real es tate while the size of the inheritance does not, cleanly separating the credit-constraint channel from the pure wealth channel. Finally, early inter vivos transfers are highly un equal in both access and impact: while they are associated with higher wealth accumulation for some households, particularly those with greater economic advantage, they are dispro portionately received by wealthier and more educated families, thereby reinforcing wealth inequality rather than offsetting it.

The remainder of the paper is organized as follows. Section 2 describes the data. Sec tion 3 analyzes the cumulative wealth channel of inheritance timing. Section 4 develops a simple model of credit constraints and provides empirical evidence by education group. Section 5 examines the implications of inheritance timing for homeownership, other real estate assets, and entrepreneurship. Section 6 explores the role of inter vivos transfers as a hedging mechanism. Section 7 concludes.

2 The Data

This study relies on household-level data from the Spanish Survey of Household Finances (EFF), a triennial (biennial after 2020) survey implemented by the Bank of Spain (Banco de España) since 2002 and covering eight waves through 2022. Each wave comprises approx imately 6,000 households drawn via stratified random sampling, with deliberate oversam pling of the top wealth quintile to ensure robust coverage of high-net-worth families. The EFF collects granular balance-sheet information (assets by category, liabilities), income and consumption flows, subjective indicators of credit constraints, and rich demographic and socio-economic covariates (age, education, occupation). Crucially, it incorporates retrospec tive modules on asset origin—purchase, inheritance, or gift—facilitating a precise analysis of intergenerational wealth transfers. Survey design, weighting, and post-2011 census recalibration are detailed in Banco de España (2024), while its alignment with the Eurosystem Household Finance and Consumption Survey supports cross-country comparability (Eu rosystem Household Finance and Consumption Network, 2013).

Inheritances and inter vivos gifts in Spain are subject to the Impuesto sobre Sucesiones y Donaciones (ISD), a transfer tax whose administration was devolved to regional govern ments (Comunidades Autónomas) in 1987. This has generated substantial cross-regiona variation in effective tax rates, exemptions, and allowances, with some regions virtually abolishing the tax for close relatives while others retain significant liability. On the civil law side, Spanish inheritance is governed by the concept of legítima: a mandatory reserved share of the estate—typically two thirds—that must pass to forced heirs (descendants, and in their absence ascendants), leaving only the remaining third (tercio de libre disposición) at the testator’s discretion. Within the reserved share, the mejora allows parents to favor one child over others up to one third of the estate. In practice, it is also common to leave assets in usufruct to the surviving spouse (usufructo vidual), which delays the effective transfer of property to children until the death of the longer-lived parent—a feature that contributes mechanically to the postponement of inheritance receipt documented in this paper. These institutional features imply that, unlike in common-law countries, Spanish heirs have lim ited ability to anticipate or accelerate the timing of bequests, making the credit-constraint channel analyzed here particularly relevant.

For each household’s first reported inheritance we extract the nominal amount and de flate it to constant 2022 euros, compute the time elapsed since receipt as the difference between the survey year and the reported year of inheritance, and recover the age at inheri tance by subtracting years since receipt from the reference person’s age at interview. These constructed variables—timing, amount, and age at receipt—constitute the core explanatory inputs in our analysis.

Table 1 summarizes the main features of the sample. Panel A reports the evolution of sample size, net wealth, income, household size, and reference-person age across waves. The pronounced divergence between mean and median net wealth points to persistent concen tration at the top of the distribution, whereas income dynamics are comparatively muted1. Panel B reports the incidence and timing of intergenerational transfers. Inheritances have become markedly more widespread—from roughly a quarter of households in 2002 to half by 2022—while being received at progressively older ages, rising from a mean of 37 years in 2002 to 45 years in 2022; the corresponding cross-sectional breakdown by age group is reported in Appendix Figure A.1, which shows that the rise is pronounced across all co horts and most visible among households aged 65 and above. Inter vivos gifts remain less common2 and are received earlier in the life cycle, consistent with a distinct role in relaxing liquidity constraints. Pooled descriptive statistics on inheritance amounts, ages, and net wealth are reported in Appendix Table A.1.

Figure 1 plots the evolution of the mean and median age at which individuals receive their first inheritance from 1970 to 2022.3 Both series display a clear upward trend: in the early 1970s the mean and median age were around 35 years, whereas by 2022 they had increased to approximately 55 years. A key institutional factor amplifying this postponement is the common practice of leaving assets in usufruct to the surviving spouse (usufructo vidual), which delays the effective transfer of property to children until the death of the longer lived parent—a margin that rises mechanically faster than individual life expectancy. This postponement of intergenerational transfers is the central motivation of the paper.4

1The divergence between mean and median net wealth reflects a persistent and widening concentration at the top of the distribution. Conde-Ruiz and García-Rodríguez (2025) document that the top 1% of Spanish households held around 13.6% of total net wealth in 2002, rising to 21.1% by 2022, while the bottom 50% maintained a marginal share of roughly 7% throughout the period. The Gini index of net wealth increased from 0.57 in 2002 to 0.69 in 2022, confirming a structural trend toward greater wealth concentration. Appendix Figures A.3, A.4 and A.5 decompose mean and median net wealth and the Gini index by age group across waves.

2see Appendix Figure A.2

3Appendix Figure A.6 replicates this plot using the largest inheritance received rather than the first.

Table 1 : EFF Sample and Intergenerational Transfers, 2002–2022

Notes: Net wealth and income in constant 2022 EUR. Ref. Age refers to the age of the reference person. Inheritances and Gifts refer to the first bequest and first inter vivos transfer reported by the household. In the earliest waves fewer response options related to inheritances are available. All statistics are computed using population weights, except the number of observations which reports the unweighted sample size. Source: Authors’ own construction based on the EFF (Banco de España, 2024).

Figure 1 : Age at First Inheritance: Survey (EFF) versus Administrative Records (Panel de Hogares).

Source: Authors’ own construction based on the EFF (Banco de España) and the Panel de Hogares (Instituto de Estudios Fiscales).

We validate the survey-based pattern against administrative records from the Span ish Inheritance and Gift Tax, compiled by the Instituto de Estudios Fiscales in the Panel de Hogares. The latest available wave (2023) covers 3.5 million individuals, offering a near population view of inheritance receipt that is not subject to self-reporting, recall, or attrition biases. Figure 1 superimposes both series; the administrative and survey-based trajectories align tightly in levels and slope, providing strong evidence that the postponement of inter generational transfers is a genuine economic phenomenon rather than a survey artifact. We nonetheless retain the EFF as our primary source because it reports inheritance amounts— essential to our identification—covers inheritances received much further back in time, and provides detailed household-level information on real assets, financial assets, and liabilities.

4Two factors explain why the roughly twenty-year increase in the age at first inheritance exceeds the rise in individual life expectancy over the same period. First, survivorship bias amplifies the early part of the series, as discussed below. Second, the relevant demographic margin is the joint survival age of the parental couple: the age at which the longer-lived parent dies rises faster than mean life expectancy, particularly when mortality improvements reduce the correlation between spousal ages at death.

To assess the quantitative relevance of survivorship bias we compute the growth rate of the age at first inheritance separately for earlier and more recent subperiods. The increase is stronger between 1970 and 1995 (1.91% annually for the mean and 1.28% for the median), but the upward trend persists between 1995 and 2022, with annual increases of 0.72% for the mean and 0.62% for the median. Truncation therefore amplifies but does not drive the documented postponement.

Figure 2 complements this analysis by showing the evolution of the mean age at first inheritance disaggregated by educational attainment.5 All groups exhibit an upward trend, but clear differences emerge: individuals with primary education tend to inherit at sig nificantly older ages—often over 60 years in recent decades—whereas those with higher education inherit earlier, with the average age remaining below 50. These patterns are consistent with systematic differences in parental fertility timing across education groups: lower-educated parents have children at younger ages but also shorter life expectancies, and the gap in childbearing age more than offsets the longevity gap, so their children inherit later.6 This educational gradient is a first indication that the timing of inheritance has heterogeneous implications for wealth accumulation across socioeconomic groups.

5Figure A.7 in the Appendix replicates this plot using the largest inheritance received.

6See Figures A.8 and A.9 in the Appendix, and the related literature in Blackburn et al. (1993), Castro (1992), and Gómez-Acebo et al. (2020).

Figure 2 : Mean Age at which the First Inheritance is Received, by Educational Level.

Source: Authors’ own construction based on the EFF.

Appendix Figures A.10 shows that bequests are heavily concentrated at the top of the net-wealth distribution: the bottom five deciles capture roughly 27% of total inherited value, the middle four about 39%, and the top decile alone more than 36%. Appendix Figure A.11 complements this picture by reporting the ratio of inherited wealth to total net wealth within each decile: inheritances exceed 300% of net wealth in the bottom decile, fall to around 79% in the second and below 50% from the fourth decile onward, and account for about 25% of net wealth in the top decile. Figures A.12 and A.13 replicate this analysis for inter vivos gifts.

3 Inheritance Timing and the Cumulative Wealth Channel

In this section, we quantify how bequest receipt and its timing shape household wealth trajectories. Earlier inheritances may yield higher wealth not only by extending the horizon for compounding returns, but also by providing upfront collateral to unlock credit. We first estimate the net wealth premium of ever having received an inheritance in subsection 3.1, comparing inheritors to non-inheritors -while controlling for demographic, socio-economic, and year fixed effects- and also analyzing the differences in the type of inheritance received.

We then exploit variation in the age at first inheritance in subsection 3.2 to identify the marginal wealth gains associated with earlier inheritance receipt. Taken together, these exercises provide clean evidence on the cumulative wealth channel, which we distinguish from the borrowing-constraint mechanism analyzed in the next section.

3.1 Inheritances, Assets Composition, and Net Wealth Outcomes

To analyze the relationship between inheritances and household wealth accumulation, we adopt an econometric strategy that distinguishes between the extensive margin of inheri tance receipt and the composition of inherited assets. This approach allows us to assess whether receiving an inheritance is associated with higher net wealth and to what extent this association depends on the type and value of assets received. All specifications include a rich set of household-level controls7 and survey-wave fixed effects.

We first estimate the following baseline specification:

\[\log (\mathrm{NetWealth} _ {i}) = \beta_ {0} + \beta_ {1} \text {InheritanceReceived} _ {i} + \boldsymbol {\beta} _ {2} \mathbf {X} _ {i} + \tau_ {t} + \varepsilon_ {i}, \tag {1}\]

where InheritanceReceived is a binary indicator equal to one if household i has ever received an inheritance, includes demographic and socioeconomic controls, and denotes survey wave fixed effects.

Table 2 reports the results. Receiving an inheritance is strongly associated with higher net wealth: inheriting households display, on average, around 300% higher net wealth than otherwise comparable non-recipient households. Age and educational attainment are also positively and significantly correlated with net wealth, consistent with standard life-cycle accumulation patterns and persistent wealth gradients by skill level.

7All regressions in this paper control for the current age of the household reference person, the educational attainment of the reference person, household income decile, the health status of the reference person, household size, and the marital status of the reference person, in addition to the variables shown in the results tables. All specifications also include survey-wave fixed effects.

Table 2 : Effect of Receiving an Inheritance on log(Net Wealth)

Notes: Dependent variable is log(Net Wealth). Survey-wave fixed effects and additional household controls are included. Significance levels: *** p < 0.1.

To explore heterogeneity across inherited assets, we restrict the sample to households that have received an inheritance and estimate:

\[\log (\mathrm{NetWealth} _ {i}) = \beta_ {0} + \beta_ {1} \operatorname{InhType} _ {i} + \beta_ {2} \log (\operatorname{InhValue} _ {i}) + \pmb {\beta} _ {3} \mathbf {X} _ {i} + \tau_ {t} + \varepsilon_ {i}, (2)\]

where Inh captures the type of inherited asset and controls for inheri tance size.

Table 3 reveals substantial heterogeneity by asset type. Inheriting other real estate assets is associated with a large and statistically significant increase in net wealth, even after controlling for the value of the inheritance. By contrast, inheritances in the form of the main residence display a smaller wealth premium than other real estate, though the coefficient remains statistically significant. The coefficient on inheritance value is positive and highly significant, indicating that larger inheritances translate into higher net wealth, but differences across asset types persist beyond valuation effects.

Table 3 : Effect of Inheritance Type and Value on log(Net Wealth)

Notes: Sample restricted to households that have received an inheritance. Survey-wave fixed effects and household controls included. Significance levels: *** 0.1.

Overall, these results indicate that inheritances are strongly associated with higher net wealth, but that this wealth premium is highly heterogeneous across asset types. In par ticular, inheritances in the form of real estate assets play a disproportionate role in driving wealth accumulation, underscoring the importance of asset composition when analyzing intergenerational wealth transfers.

3.2 Marginal Wealth Gains from Early Bequests

We next assess how the timing of the first inheritance shapes household net wealth, beyond the bequest’s size and standard life-cycle accumulation. Disentangling the effect of timing from mere time-on-book exposure is crucial, since earlier transfers extend the horizon for compounding returns and collateral use, whereas later inheritances face a shorter accumu lation window. To isolate this mechanism, we estimate a quadratic specification in the age at inheritance, controlling for bequest size, socio-economic characteristics (including age at the time of the survey), and survey-wave fixed effects.

\[\log (\mathrm{NetWealth} _ {i}) = \alpha_ {0} + \alpha_ {1} \mathrm{AgeAtInheritance} _ {i} + \alpha_ {2} \mathrm{AgeAtInheritance} _ {i} ^ {2} + \alpha_ {3} \log (\mathrm{InhValue} _ {i})\]

\[+ \alpha_ {4} \mathbf {X} _ {i} + \tau_ {t} + \varepsilon_ {i}, \tag {3}\]

where AgeAtInheritance denotes the recipient’s age at the time of the first inheritance. The linear term captures the marginal effect of delaying receipt by one year, while the squared term allows for nonlinearities in the timing channel. The logarithm of the inher ited amount controls for bequest size, and includes standard household-level covariates. Survey-wave fixed effects absorb common macroeconomic shocks.

It is important to clarify that our specification does not attempt to separately identify a purely mechanical compounding effect from a behavioral timing effect. Conditional on current age and inheritance size, the coefficient on Age at Inheritance captures the total marginal cost of delaying inheritance receipt by one year. This total effect necessarily includes the loss of an additional year of compounding, as well as any impact operating through credit constraints or life-cycle decisions. The subsequent heterogeneity analysis by education group provides indirect evidence on the role of borrowing constraints, rather than mechanically netting out the compounding component from Equation 3.

Table 4 reports the results for the full sample. Delaying the receipt of an inheritance is associated with a statistically significant and economically meaningful reduction in net wealth. Each additional year of delay is associated with approximately 2.25% lower net wealth, while the positive quadratic term indicates that this association attenuates at older ages.

The magnitude of the timing effect is substantial. A delay of ten years in receiving the first inheritance is associated with a reduction in net wealth of roughly 22%, highlighting that the economic cost of late inheritances can rival or exceed the impact of many conventiona determinants of wealth. By contrast, the size of the transfer plays a comparatively modest role: a 10% increase in the inherited amount raises net wealth by less than 3%.

To allow for nonlinearities, we include a quadratic term in the age at inheritance. The estimates imply a turning point at approximately 56 years, where the marginal penalty of delaying an inheritance vanishes.8 This pattern indicates that the wealth penalty of delayed inheritances is concentrated in early and mid-adulthood, when capital injections have the greatest scope to accumulate and to relax borrowing constraints.

Table 4 : Effect of Age at Inheritance on log(Net Wealth)

Notes: Dependent variable is log(Net Wealth). All regressions include household-level controls and survey-wave fixed effects. Standard errors are clustered at the household level. Significance levels: *** .

In the next subsection, we explore whether this timing penalty differs systematically across educational groups, first theoretically and then empirically, shedding light on the role of human capital and credit constraints in shaping the long-run returns to intergenerational transfers.

4 Inheritance Timing and the Borrowing Constraint Channel

4.1 A Stylized Two-Period Model of Credit Constraints and Bequest Timing

The model below is intentionally stylized. Its purpose is to organize the mechanism that differentiates the credit-access channel from alternative interpretations—such as pure com pounding or income effects—rather than to generate quantitative predictions calibrated to the data. The qualitative asymmetry it produces between constrained and unconstrained households motivates the heterogeneity analysis by education in Section 4.2.

8Formally, the marginal effect is given by . Setting this expression to zero yields the critical age , which in our estimates equals 56.25 years. Intuitively, beyond this age the remaining compounding horizon is short enough that delaying receipt by one additional year generates negligible wealth losses.

Credit constraints generate heterogeneity in investment returns across households. Early contributions established that liquidity limits hinder consumption smoothing (Zeldes, 1989) and trigger sharp responses to uninsured income shocks (Townsend, 1994), while collateral based borrowing limits became workhorse mechanisms to explain wealth dispersion (Huggett, 1993; Aiyagari, 1994) and credit cycle dynamics (Kiyotaki and Moore, 1997). We build on this tradition by introducing bequest timing as a life-cycle liquidity shock that differentially relaxes borrowing constraints across households.

A central implication of this literature is the existence of a collateral threshold be low which households are effectively excluded from high-return investment opportunities (Townsend, 1994; Huggett, 1993; Aiyagari, 1994). To capture this heterogeneity in a tractable two-period setting, we posit a minimum collateral cuto! such that

\[y _ {L} < m _ {\mathrm{min}} \leq y _ {H},\]

implying that “low-ability” households (i = L) cannot access the high-return technology un less they receive an up-front transfer, while “high-ability” households (i = H) always qualify. Educational attainment serves as the empirical proxy for this distinction: households with lower education tend to face tighter borrowing constraints and have fewer alternative sources of self-financing, consistent with evidence linking human capital to credit access (Lochner and Monge-Naranjo, 2011; Popov, 2014; Mogstad, 2017; Pfeffer, 2018).

Time is indexed by . In period 1, each household receives labor income with labor income is zero in period 2. A lump-sum bequest B arrives either at the beginning of period 1 (“early bequest”) or at the beginning of period 2 (“late bequest”). Preferences are

\[U _ {i} = \log (c _ {1, i}) + \beta \log (c _ {2, i}), \qquad \beta \in (0, 1).\]

Households choose period-1 consumption and savings subject to

\[c _ {1, i} + s _ {i} = m _ {1, i}, \qquad m _ {1, i} = y _ {i} + B \cdot \mathbf {1} \{\text { early bequest } \}.\]

Savings earn gross return 1+R for households with access to the productive technology and

for those excluded from it, with . Period-2 consumption is therefore

\[c _ {2, i} = (1 + \rho_ {i}) s _ {i} + B \cdot {\bf 1} \{\mathrm{latebequest} \},\]

where reflects credit access. Substituting the first-order condition for the household’s problem yields the log-utility allocation

\[c _ {2, i} ^ {*} = \frac {\beta}{1 + \beta} \Big [ (1 + \rho_ {i}) m _ {1, i} + B \cdot \mathbf {1} \{\mathrm{latebequest} \} \Big ].\]

Defining the welfare gain from early rather than late receipt as , the model delivers a sharp asymmetry across household types. For high-ability households, credit access is unconstrained regardless of timing in both scenarios), so

\[\Delta c _ {2, H} = \frac {\beta}{1 + \beta} R B.\]

The gain reflects exclusively the additional period of returns earned by investing the bequest sooner—the investment channel. For low-ability households, by contrast, timing determines credit access: in the late-bequest scenario they earn return in the early-bequest scenario the up-front transfer unlocks the high-return technology. The implied welfare gain is

\[\Delta c _ {2, L} = \frac {\beta}{1 + \beta} \Big [ (R - r) y _ {L} + R B \Big ].\]

This expression decomposes into two components. The term captures a creditaccess channel: early liquidity allows low-ability households to deploy their labor income at the higher return R rather than the safe rate r. The term ε is the investment channel shared with unconstrained households. For tractability, the model focuses on the extensive margin of credit access and abstracts from intensive-margin leverage, thereby isolating the role of eligibility and timing rather than optimal leverage. The key asymmetry between the two household types is summarized in the following proposition.

Proposition 1 The welfare gain from early inheritance receipt is strictly larger for credit-

constrained households than for unconstrained ones:

\[\Delta c _ {2, L} - \Delta c _ {2, H} = \frac {\beta}{1 + \beta} (R - r) y _ {L} > 0.\]

The gap is increasing in the return spread and in the labor income of the constrained household .

The proof follows directly from substitution and the assumptions and . The proposition formalizes why inheritance timing is particularly consequential for households facing tighter borrowing constraints: beyond the investment channel available to all recip ients, early bequests unlock a credit-access channel that is exclusive to constrained house holds. Since educational attainment proxies for in the model—more educated households face fewer borrowing frictions and have greater self-financing capacity—the framework pre dicts that the timing penalty should be largest among households with lower education. We test this hypothesis directly in Section 4.2, which provides stratified estimates by ed ucational group, and Section 5, which examines the implications of inheritance timing for homeownership, entrepreneurship, and real estate investment.

While the model treats bequest timing as exogenous, sufficiently wealthy parents may in principle shift resources forward through discretionary inter vivos gifts (Cox, 1987; Hurd et al., 2011), partially offsetting the costs of delayed inheritance receipt for some households. We examine this hedging mechanism and its distributional implications in Section 6.

4.2 Empirical Evidence on Early Inheritances and Credit Constraints by Skill Level

We next examine whether the wealth premium associated with inheritance timing varies systematically across educational groups. To this end, we stratify the sample by the ref erence person’s highest completed level of education—primary, secondary, or higher—and re-estimate Equation 3 separately for each subgroup. This strategy allows us to assess how differences in human capital and credit access shape the long-run returns to intergenera tional transfers. If early inheritances primarily operate by relaxing borrowing constraints, their timing should matter most for lower-educated households, who are more likely to face

binding credit constraints9.

Table 5 : Effect of Age at Inheritance on log(Net Wealth), by Educational Level

Notes: Dependent variable is log(Net Wealth). All regressions include household-level controls and survey-wave fixed effects. Standard errors are clustered at the household level. Significance levels: *** p < 0.01, ** p < 0.05, * p < 0.1.

Table 5 documents pronounced heterogeneity in the timing effect of inheritances across educational groups10. Among households with primary education, delaying inheritance re ceipt is associated with a large and statistically significant wealth penalty: a one-year in crease in the age at inheritance is linked to approximately 3.85% lower net wealth. The positive and significant quadratic term indicates a pronounced nonlinear profile, with the relationship strongest at younger ages.

For households with secondary education, the timing penalty remains sizable but less steep. A one-year delay corresponds to roughly 2% lower net wealth, and the curvature of the profile suggests that the strength of this relationship gradually declines over the life cycle. In contrast, for highly educated households, neither the linear nor the quadratic timing terms are statistically significant, indicating that the age at which inheritances are received does not systematically affect long-run wealth outcomes for this group.

By comparison, the coefficient on inheritance value is positive and highly significant across all educational categories, with remarkably similar magnitudes. This pattern implies that while the size of the transfer matters for wealth accumulation across all educationa groups, differences in wealth trajectories by education are driven primarily by variation in inheritance timing rather than by differences in transfer size, as the estimated effect of inheritance value is similar across groups. The timing channel, by contrast, operates almost exclusively among lower- and middle-educated households.

9To further assess the credit-constraints interpretation, we re-estimate the specifications in Table 5 conditioning on quartiles of inheritance size, so that we compare households receiving transfers of similar magnitude at different stages of the life cycle. The results, reported in Table B.1 in the Appendix, show that the timing penalty remains significant for primary-educated households in the intermediate quartiles of the inheritance distribution (Q2–Q3) and for secondary-educated households in the upper-intermediate quartile (Q3), while it is statistically negligible across all quartiles for highly educated households. This pattern is consistent with differences in credit access across education groups rather than with differential parental wealth composition.

10All results reported hereafter by educational attainment have also been replicated using income-based classifications. The main findings remain qualitatively and quantitatively unchanged. See the Appendix for details.

Figure 3 illustrates these dynamics11. The predicted wealth profiles reveal steep declines in net wealth as inheritance receipt is delayed among low-educated households, a more mod erate slope for those with secondary education, and a nearly flat profile for highly educated households. These results are consistent with the presence of binding credit constraints among lower-skilled households, for whom early capital injections relax borrowing limits precisely when liquidity is most valuable. By contrast, higher-educated households appear largely insulated from timing effects, likely reflecting greater access to credit and higher self-financing capacity.

11The heterogeneity in the timing penalty across education groups is formally validated by a joint test of the interactions between inheritance timing and education dummies, which rejects equality of coefficients across all three groups . Pairwise tests confirm that the timing penalty is statistically different between primary and higher educated households both in the linear term and jointly with the quadratic , and between secondary and higher educated households at the 10% level . The difference between primary and secondary is significant at the 10% level when both timing terms are considered jointly .

Figure 3 : Relationship Between Age at Inheritance Receipt and Net Wealth (in log), by Educational Level

Source: Authors’ own construction based on the regression results. Note: For highly educated households, neither the linear nor the quadratic timing coefficient is statistically significant.

A natural concern is that the heterogeneity-by-education pattern documented in Table 5 could reflect systematic differences in parental socioeconomic composition rather than differ ential credit access. Specifically, highly educated heirs might descend from longer-lived and wealthier parents, producing observationally equivalent differences in the timing and size of inheritance receipt. Two pieces of evidence argue against this interpretation. First, all spec ifications control for the logarithm of inheritance value, absorbing the wealth-composition dimension of parental socioeconomic status. Second, and more fundamentally, Figure 2 shows that the empirical gradient in age at inheritance runs in the direction opposite to what a pure parental-longevity selection story would predict: lower-educated heirs inherit substantially later, not earlier, because the educational gradient in fertility timing more than offsets the gradient in parental longevity (see also Appendix Figures A.8 and A.9). The absence of a timing penalty among highly educated households, who are empirically more exposed to early inheritance receipt, is therefore difficult to reconcile with a selec tion narrative and is consistent with differential credit access across education groups. The stratification by inheritance-size quartiles reported in Table B.1 in the Appendix further re inforces this interpretation by showing that the timing penalty persists for primary-educated households within intermediate quartiles of bequest magnitude.

Beyond wealth accumulation, these findings suggest that early inheritances expand households’ feasible choice sets in an unequal manner. By relaxing collateral constraints at younger ages, early transfers open access to credit-financed opportunities—such as en trepreneurship, homeownership, and family formation—that are otherwise constrained for lower-skilled households. In the next section, we test these mechanisms directly by exam ining how inheritance timing affects business ownership, housing outcomes, and fertility decisions.

5 Early Inheritance and Key Economic Decisions

While the previous section showed that receiving an inheritance earlier in life generates substantial long-run wealth gains—particularly for households facing tighter credit con straints—wealth accumulation is only one dimension through which intergenerational trans fers may shape economic outcomes. Early inheritances may also influence a set of discrete and highly consequential economic decisions that determine households’ trajectories over the life cycle. In particular, by relaxing liquidity and collateral constraints at younger ages, early transfers may facilitate access to homeownership, enable investment in additional real estate assets, and lower barriers to entrepreneurial activity.

To examine these channels, this section analyzes how the timing of inheritance receipt affects the likelihood of engaging in three key economic outcomes12: owning the main resi dence, owning additional real estate properties, and operating a business. Unlike net wealth, these outcomes reflect discrete choices that are often characterized by high entry costs and credit requirements, making them especially sensitive to the availability of early capital injections.

We estimate the following empirical specification:

\[Y _ {i} = \alpha_ {0} + \alpha_ {1} \text {AgeAtInheritance} _ {i} + \alpha_ {2} \log (\text {InhValue} _ {i}) + \alpha_ {3} \mathbf {X} _ {i} + \tau_ {t} + \varepsilon_ {i}, \tag {4}\]

12Because the data link current outcomes to the timing of inheritance receipt, the estimates should be interpreted as documenting conditional long-run associations rather than strictly causal inheritance effects.

where is a binary indicator capturing the probability of the three outcomes of interest: (i) homeownership of the primary residence, (ii) ownership of other real estate properties, or (iii) business ownership13. The coefficient measures how the probability of each outcome varies with the age at which the inheritance is received14, holding constant the size of the transfer and household characteristics. The logarithm of the inherited amount controls for wealth effects arising purely from inheritance size, while includes standard demographic and socioeconomic controls. Survey-wave fixed effects, , absorb common macroeconomic conditions and institutional changes over time.

By focusing on these outcomes, the analysis moves beyond wealth levels to assess whether early inheritances translate into tangible economic choices that shape households’ productive capacity and long-term economic security. The following subsections present the estimated effects for each outcome in turn.

5.1 Homeownership of the Primary Residence

Homeownership is a central component of household economic security and a key channel through which wealth is accumulated over the life cycle. However, access to owner-occupied housing typically requires substantial upfront liquidity, most notably to meet down-payment requirements, making entry into homeownership particularly sensitive to credit constraints. Early inheritances may therefore play a critical role by providing liquid resources at a stage of the life cycle when households are most likely to face binding borrowing limits.

To examine this channel, we estimate probit models15 in which the dependent variable is an indicator for owning the main residence. The key explanatory variables are the age at which the inheritance is received and the logarithm of the inherited amount, controlling for household characteristics and survey-wave fixed effects. Table 6 reports average margina effects for the full sample and separately by educational attainment.

13Although not a purely economic outcome, we also examine the relationship between the timing of inheritance receipt and fertility outcomes. See Section C in the Appendix for details.

14We do not include a quadratic term in age at inheritance in these specifications. When tested, the squared term is not statistically significant for any of the outcomes considered. This is also consistent with the underlying theoretical framework, as these economic decisions are predominantly taken early in the life cycle, implying a monotonic—rather than nonlinear—relationship between inheritance timing and discrete economic choices.

15Although the tables report results from probit models, we have also estimated logit specifications as a robustness check. The sign and statistical significance of the marginal effects remain virtually unchanged, confirming the robustness of our findings.

Table 6 : Marginal Effects on the Probability of Homeownership

Notes: Average marginal effects from probit models. All regressions include householdlevel controls and survey-wave fixed effects. Robust standard errors clustered at the household level in parentheses. Significance levels: *** .

The results indicate that the timing of inheritance receipt is a statistically significant determinant of homeownership, but only for households with low and intermediate levels of education. In the full sample, households who receive an inheritance one year earlier exhibit about a 0.12 percentage point higher probability of owning their main residence. This effect is economically meaningful when accumulated over several years and is primarily driven by households with primary and secondary education.

Among primary-educated households, a one-year delay in inheritance receipt is associ ated with approximately a 0.11 percentage point lower probability of homeownership, while the relationship is slightly stronger for those with secondary education (0.15 percentage points). By contrast, the timing coefficient for highly educated households is small and statistically insignificant, suggesting that inheritance timing plays a limited role in shaping homeownership outcomes for this group.

In contrast to timing, the scale effect of inheritances is positive and statistically signifi cant across all educational groups. A 10% increase in inheritance value raises the probability of homeownership by roughly 0.14 percentage points in the full sample, with remarkably sim ilar magnitudes across education levels. This pattern suggests that while larger inheritances facilitate housing acquisition for all households, early inheritances provide an additional and distinct advantage only for those facing tighter borrowing constraints.

Overall, these findings reinforce the central role of inheritance timing—rather than in heritance size—in easing access to homeownership among credit-constrained households.

Early inheritances help overcome down-payment barriers precisely when households are most likely to be liquidity constrained, whereas highly educated households appear largely able to compensate for delayed transfers through alternative sources of financing.

5.2 Ownership of Additional Real Estate Assets

Beyond access to the primary residence, the acquisition of additional real estate properties represents a distinct margin of economic behavior closely linked to long-run wealth accu mulation through capital gains and rental income. Entry into this segment of the housing market typically requires substantial upfront capital and sustained access to credit, making it particularly sensitive to the timing and size of intergenerational transfers.

We estimate probit models in which the dependent variable indicates ownership of real estate assets other than the primary residence, controlling for inheritance timing, inheritance value, household characteristics, and survey-wave fixed effects. Table 7 reports average marginal effects for the full sample and by educational attainment.

Table 7 : Marginal Effects on the Probability of Owning Other Real Estate

Notes: Average marginal effects from probit models. All regressions include householdlevel controls and survey-wave fixed effects. Robust standard errors clustered at the household level in parentheses. Significance levels: ***

The results differ significantly from those for primary homeownership. In the full sample, receiving an inheritance earlier has a positive impact on the probability of owning additiona real estate assets, but this effect is almost entirely driven by households with primary edu cation. For this group, a one-year delay in inheritance receipt is linked with a reduction in the probability of owning other real estate by more than 0.3 percentage points.

By contrast, inheritance timing is not statistically significant for households with secondary or higher education. For these groups, the size of the inheritance plays a much more prominent role, with larger inheritances substantially increasing the likelihood of owning additional real estate. This asymmetry highlights that other real estate ownership is the margin along which inheritance timing generates the largest wealth disparities: early trans fers enable low-educated households to access real estate investment opportunities that are otherwise out of reach, while more educated households rely primarily on inheritance size and alternative financing channels.

5.3 Entrepreneurial Entry

Entrepreneurial activity is highly sensitive to liquidity and credit access, as starting a business typically requires upfront capital and involves substantial risk. Early inheritances can therefore ease financial constraints at a critical stage of the life cycle, facilitating entry into entrepreneurship. To quantify this channel, we estimate probit models for the probability of entrepreneurship as a function of inheritance timing and value, controlling for household characteristics and survey-wave fixed effects.16 Table 8 reports average marginal effects for the full sample and by educational attainment.

Table 8 : Marginal Effects on the Probability of Entrepreneurship

Notes: Average marginal effects from probit models. All regressions include householdlevel controls and survey-wave fixed effects. Robust standard errors clustered at the household level in parentheses. Significance levels: ***

The estimates point to a clear timing effect on entrepreneurial entry. In the full sample, receiving an inheritance one year earlier corresponds to roughly a 0.20 percentage point higher probability of starting a business. This effect is present across all educational groups, but it is strongest among individuals with secondary education and remains statistically significant even among highly educated households17.

16Results are robust to alternative logit specifications.

In contrast, the role of inheritance size is more uneven. Larger inheritances increase entrepreneurial entry mainly among secondary-educated households, while the scale effect is weaker and statistically insignificant for the remaining groups. Taken together, these results suggest that early inheritances facilitate entrepreneurship primarily by improving liquidity at the right moment, rather than by providing large amounts of capital.

6 Inter Vivos Transfers as a Hedging Mechanism

The previous sections have documented that demographic aging systematically delays in heritance receipt, imposing a substantial timing penalty on wealth accumulation and life cycle decisions—particularly among households facing binding credit constraints. A natural question is whether families can anticipate these costs by shifting resources forward through discretionary inter vivos transfers. We extend the benchmark model of Section 4.2 to ad dress this question. Unlike inheritances, which occur involuntarily at death, inter vivos transfers reflect discretionary donor decisions subject to life-cycle and resource constraints (Cox, 1987; Hurd et al., 2011).

Let B denote the total amount transferred across generations. Parents may choose to transfer a portion in period 1 as an inter vivos gift, while the remaining amount is transferred as a bequest in period 2. From the child’s perspective, a gift is equivalent to an early bequest: market resources in period 1 become , and the child’s indirect utility is increasing in g, with a steeper slope for households facing binding credit frictions (Guiso and Jappelli, 2002). Early transfers are therefore particularly valuable for liquidity-constrained children.

Parents have resources W and must retain at least ¯c to finance their own consumption and retirement needs (Modigliani, 1988; Hurd et al., 2011). This generates the feasibility

17The timing effect on entrepreneurship is statistically significant across all education groups (Table 8), unlike the patterns documented for housing outcomes. This is consistent with the nature of entrepreneurial finance: banks typically require physical collateral to fund new ventures regardless of the borrower’s edu cational background (Kiyotaki and Moore, 1997; Evans and Jovanovic, 1989), making entrepreneurship a margin where credit constraints bind more broadly across the socioeconomic spectrum.

constraint

\[0 \leq g \leq \min \bigl \{B, \max (W - \bar {c}, 0) \bigr \}.\]

Parents choose g to maximize subject to this constraint. If , then and any transfer must take the form of a late bequest. The model therefore delivers two empirical implications that we take to the data. First, early inter vivos transfers should be concentrated among households at the top of the wealth and education distributions, since only families with can shift transfers forward (De Nardi, 2004; Nishiyama, 2002). Second, the hedging channel is inherently unequal: the households that would benefit most from early liquidity—those with binding credit constraints—are precisely those least likely to have access to it. We examine both predictions in turn.

Figure 4 shows the diverging trajectories of inheritances and inter vivos gifts over the sample period. The average age at first inheritance increased from 37.4 to 44.7 years between 2002 and 2022, while the age at first gift receipt rose more modestly over the same period. As a result, gifts provide access to wealth substantially earlier in the life cycle than inheritances, suggesting that they could in principle serve as an anticipatory liquidity channel. This timing gap is consistent with recent contributions emphasizing that the timing of intergenerational transfers is a key determinant of their economic relevance (Boileau and Sturrock, 2023; Pasini et al., 2024). Importantly, however, this does not imply that gifts fully compensate for delayed inheritances: the first condition for the hedging channel to operate is satisfied, but the second—that access be sufficiently widespread—remains to be examined.

Figure 4 : Mean and median age at first inheritance and first gift receipt.

Source: Authors’ own construction based on the EFF.

The availability of this channel is highly uneven. Figure 5 documents a pronounced educational gradient in both the timing and the incidence of inter vivos transfers. Lower educated households not only receive gifts later in life, but do so far less frequently than those with tertiary education. Since educational attainment proxies for long-run family resources, access to credit, and the capacity to deploy early liquidity productively (Corak, 2013), this pattern indicates that anticipatory wealth transmission is selective: it reinforces the advantages of already privileged households rather than offsetting the timing penalty faced by more constrained ones.

Figure 5 : Inter vivos gifts by educational attainment of the recipient. (a) Mean age at first gift receipt.

(b) Share receiving an inter vivos gift. Source: Authors’ own construction based on the EFF.

This selectivity extends to family background. Figure 6 relates the incidence of gifts to parental occupational status, classified by skill level following the International Standard Classification of Occupations (ISCO) and used as a proxy for parental human capital. House holds whose parents hold high-skilled occupations are substantially more likely to receive inter vivos transfers—and, in particular, transfers occurring before inheritance receipt— than those whose parents are in low-skilled occupations.18 These patterns are consistent with strong intergenerational persistence in socioeconomic status and with the notion that only a subset of families has the economic capacity to anticipate wealth transfers (Corak, 2013; Elinder et al., 2018).

18The Appendix Figure D.1 reports complementary evidence on the average value of gifts by parental occupational status. Because gift amounts are highly skewed and sample sizes within cells are limited, these estimates are more volatile; the overall patterns are nonetheless consistent with the incidence results shown here.

Figure 6 : Incidence of inter vivos gifts by parental occupational status.

Notes: Parental occupation is classified by skill level following ISCO and used as a proxy for socioeconomic background. The sample is restricted to households in which both parents (or parents-in-law) fall into the same skill category. The horizontal axis reports EFF survey waves rather than years of gift receipt to ensure sufficient sample size within each cell.

Source: Authors’ own construction based on the EFF.

To further characterize the distributional nature of inter vivos transfers, we estimate a set of regressions relating their incidence and magnitude to households’ position in the wealth distribution, conditional on demographic and socioeconomic characteristics. Because the EFF does not provide direct information on donors’ wealth or education, the analysis focuses on recipient characteristics. In a context of limited intergenerational mobility, house holds’ position in the wealth distribution is informative—though imperfectly—about long run family resources and, by extension, about the economic capacity of potential donors (Corak, 2013). We consider three outcomes: the probability of receiving an inter vivos transfer, the probability of receiving both an inheritance and a gift, and the logarithm of the gift value:

\[\left( \begin{array}{c} \operatorname * {P r} (\mathrm{Gift} _ {i t} = 1) \\ \operatorname * {P r} (\text {BothInheritanceGift} _ {i t} = 1) \\ \log (\text {GiftAmount} _ {i t}) \end{array} \right) = \beta_ {0} + \beta_ {1} \text {WealthDecile} _ {i} + \mathbf {X} _ {i t} ^ {\prime} \boldsymbol {\beta} _ {2} + \tau_ {t} + \varepsilon_ {i t}, \tag {5}\]

where WealthDecile denotes the household’s position in the net wealth distribution, in cludes demographic and socioeconomic controls, and are survey-wave fixed effects. These regressions are descriptive in nature and are not intended to identify causal effects.

Figure 7 summarizes the results. The probability of receiving an inter vivos transfer increases sharply and monotonically with wealth, wealthier households are substantially more likely to combine gifts and inheritances, and the value of gifts rises steeply along the wealth distribution. These patterns are consistent with existing evidence that intergenera tional transfers are highly concentrated at the top of the wealth distribution (Elinder et al., 2018; De Nardi, 2004), and confirm the first theoretical prediction: the capacity to antic ipate transfers is concentrated among wealthier families, precisely because the feasibility constraint W > c¯ is binding for most households.19

19Appendix Figure D.2 confirms that these patterns are robust to using income deciles rather than wealth deciles to characterize households’ position in the distribution.

Figure 7 : Inter-vivos Transfers and Net Wealth Deciles (a) Probability of Receiving an Inter-vivos Gift

(b) Probability of Receiving Both an Inheritance and a Gift

(c) Predicted Log(Value of Early Gift)

Notes: Each panel reports predicted values by net wealth decile, holding covariates at their means. Shaded areas correspond to 95% confidence intervals. All specifications control for demographic characteristics, education, and survey wave fixed effects.

Source: Authors’ own construction based on the EFF.

Finally, Figure 8 shows that the concentration of inter vivos transfers among the wealthy is even more pronounced when attention is restricted to transfers that are truly anticipatory. Defining an early gift as an inter vivos transfer received prior to the first inheritance and at or before age 45—the threshold below which transfers can plausibly relax credit constraints at critical stages of the life cycle (Boileau and Sturrock, 2023; Pasini et al., 2024)—the predicted probability of receiving such a transfer increases sharply and monotonically with wealth, remaining close to zero throughout the bottom half of the distribution and rising steeply only in the top two deciles.

Taken together, these findings confirm both testable predictions derived from the model.

Figure 8 : Predicted Probability of Receiving an Early Inter-Vivos Gift

Notes: Predicted probabilities by net wealth decile, holding covariates at their means. Shaded areas correspond to 95% confidence intervals. Source: Authors’ own construction based on the EFF.

Early inter vivos transfers are overwhelmingly concentrated among households at the top of the wealth and education distributions, and the hedging channel is inherently unequal: the households that would benefit most from early liquidity—those facing binding credit constraints—are precisely those least likely to have access to it. Rather than offsetting the distributional consequences of delayed inheritance receipt, inter vivos transfers reinforce them. Demographic aging therefore amplifies wealth inequality through a dual mechanism: by delaying inheritances for everyone, and by concentrating the only available hedge against that delay among those who are already advantaged.

7 Conclusions

This paper shows that demographic aging reshapes intergenerational wealth transmission not only by increasing the volume of wealth transferred across generations, but also by systematically delaying the timing at which these transfers are received. As rising life ex pectancy postpones inheritances to later stages of the life cycle, their capacity to relax credit constraints, influence major economic decisions, and foster long-run wealth accumulation is substantially weakened. Importantly, this timing effect is highly uneven across house holds, giving rise to a dual mechanism through which demographic aging amplifies wealth inequality.

Our analysis delivers four main insights. First, inheritances are strongly associated with higher household net wealth, but their economic impact depends critically on when they are received. Delays in inheritance receipt are linked to systematically lower long-run wealth, with the largest penalties arising when transfers occur during early and mid-adulthood. Timing therefore emerges as a central dimension of intergenerational transfers, distinct from their magnitude.

Second, the timing penalty is strongly heterogeneous and closely linked to credit con straints. Delayed inheritances impose sizable costs on households with low and medium education—groups that face tighter borrowing constraints and rely more heavily on early access to capital—while the effect is small and statistically negligible for highly educated households. This pattern indicates that the economic consequences of delayed transfers are concentrated precisely among those for whom liquidity is most valuable.

Third, inheritance timing shapes a range of key life-cycle decisions with long-run con sequences. Earlier transfers are associated with higher probabilities of entrepreneurship, greater access to homeownership and other real estate assets, and earlier family formation. These effects are consistently stronger among less educated households, underscoring the role of early wealth in relaxing binding financial constraints at critical stages of the life cycle and shaping long-term economic trajectories. This pattern is especially stark for in vestment in additional real estate among primary-educated households, for whom the timing of the transfer predicts ownership while the size of the inheritance does not—a contrast that separates the credit-constraint channel from the pure wealth channel.

Fourth, and central to the distributional implications of aging, inter vivos transfers pro vide an incomplete and highly uneven buffer against delayed inheritances. While early gifts are associated with higher levels of wealth accumulation, access to this channel is strongly concentrated among wealthier and more educated families. Both the likelihood of receiving early transfers and the magnitude of gifts increase sharply with household economic advan tage, implying that anticipatory transfers disproportionately benefit those who are already better positioned to accumulate and deploy capital. Rather than equalizing opportunities, inter vivos transfers therefore amplify existing disparities in wealth accumulation and rein force inequalities in the timing and effectiveness of intergenerational wealth transmission.

Taken together, these findings point to a dual inequality channel induced by demographic aging. On the one hand, delayed inheritances weaken the transformative role of intergen erational transfers for credit-constrained households. On the other hand, the capacity to anticipate transfers through early gifts allows more advantaged families to shield their de scendants from these losses. As a result, the timing of intergenerational transfers emerges as a key mechanism through which economic advantage persists across generations in aging societies.

These results have broader implications for the study of wealth inequality and inter generational mobility. They suggest that analyses of intergenerational transfers may be incomplete if they focus exclusively on the size of transfers, abstracting from when wealth is received over the life cycle. From a policy perspective, our findings suggest that the life-cycle stage at which wealth arrives is as consequential as its magnitude for economic and distributional outcomes. This points to a role for policy instruments that are sensitive to the timing of transfers, not merely their size. For instance, inheritance tax schedules that vary with the age of the recipient at the time of receipt could better account for the unequal economic impact of late bequests across socioeconomic groups. Similarly, fiscal incentives for early inter vivos transfers—conditional on donor wealth to avoid reinforcing existing concentration—could partially offset the liquidity penalties documented here for credit-constrained households. We leave the design and welfare evaluation of such instru ments to future research.

While our analysis focuses on data from Spain, future research could examine whether similar timing effects arise in other institutional contexts. The Spanish Survey of Household Finances is part of the harmonized Household Finance and Consumption Survey (HFCS), which covers more than twenty European countries. Extending this framework to a cross country setting would allow systematic comparisons across inheritance regimes, credit market structures, and demographic dynamics, helping to establish whether the timing premium documented here is a general feature of aging societies or depends on specific

institutional environments.

References

1. Aiyagari, S. R. (1994). Uninsured idiosyncratic risk and aggregate saving. American Economic Review, 84(3):659–684.
2. Andersen, S. and Nielsen, K. M. (2011). Participation constraints in the stock market: Evidence from unexpected inheritance due to sudden death. Review of Financial Studies, 24(5):1667–1697.
3. Banco de España (2024). Spanish survey of household finances (EFF) 2022: methods, results and changes since 2020. Documentos Ocasionales 2413, Banco de España, Madrid.
4. Black, S. E., Devereux, P. J., Landaud, F., and Salvanes, K. G. (2025). The (un) importance of inheritance. Journal of the European Economic Association, 23(3):1060–1094.
5. Blackburn, M. L., Bloom, D. E., and Neumark, D. (1993). Fertility timing, wages, and human capital. Journal of Population Economics, 6(1):1–30.
6. Boileau, B. and Sturrock, D. (2023). What drives the timing of inter-vivos transfers? Technical report, IFS Working Papers.
7. Boserup, S. H., Kopczuk, W., and Kreiner, C. T. (2016). The role of bequests in shaping wealth inequality: evidence from Danish wealth records. American Economic Review, 106(5):656–661.
8. Castro, T. (1992). Delayed childbearing in contemporary spain: trends and differentials. European Journal of Population/revue Européenne De Démographie, 8(3):217–246.
9. Conde-Ruiz, J. I. and García-Rodríguez, F. (2025). Evolución de la riqueza de las familias en España (2002–2022): Niveles, composición y fractura generacional. Technical report, FEDEA.
10. Corak, M. (2013). Income inequality, equality of opportunity, and intergenerational mobility. Journal of Economic Perspectives, 27(3):79–102.
11. Cox, D. (1987). Motives for private income transfers. Journal of Political Economy, 95(3):508–546.
12. De Nardi, M. (2004). Wealth inequality and intergenerational links. The Review of Economic Studies, 71(3):743–768.
13. Delgado, M., Meil, G., and Zamora-López, F. (2008). Spain: Short on children and short on family policies. Demographic Research, 19(27).
14. Druedahl, J. and Martinello, A. (2022). Long-run saving dynamics: Evidence from unexpected inheritances. The Review of Economics and Statistics, 104(5):1079–1095.
15. Elinder, M., Erixson, O., and Ohlsson, H. (2012). The impact of inheritances on heirs’ labor and capital income. The B.E. Journal of Economic Analysis & Policy, 12(1):1–37.
16. Elinder, M., Erixson, O., and Waldenstr¨om, D. (2018). Inheritance and wealth inequality: Evidence from population registers. Journal of Public Economics, 165:17–30.
17. Engelhardt, G. V. and Mayer, C. J. (1998). Intergenerational transfers, borrowing constraints, and saving behavior: Evidence from the housing market. Journal of Urban Economics, 44(1):135– 157.
18. Eurosystem Household Finance and Consumption Network (2013). The Eurosystem household finance and consumption survey-methodological report. Technical report, ECB Statistics Paper.
19. Evans, D. S. and Jovanovic, B. (1989). An estimated model of entrepreneurial choice under liquidity constraints. Journal of Political Economy, 97(4):808–827.
20. Gale, W. G. and Scholz, J. K. (1994). Intergenerational transfers and the accumulation of wealth. Journal of Economic Perspectives, 8(4):145–160.
21. Garbinti, B. and Georges-Kot, S. (2019). On the retirement effect of inheritance: heterogeneity and the role of risk aversion. Working paper.
22. Gómez-Acebo, I., Dierssen-Sotos, T., Palazuelos, C., Castaño-Vinyals, G., Pérez-Gómez, B., Ami ano, P., Fernández-Villa, T., Ardanaz, E., Suarez-Calleja, C., Alguacil, J., et al. (2020). Changes in individual and contextual socio-economic level influence on reproductive behavior in spanish women in the mcc-spain study. BMC Women’s Health, 20(1):72.
23. Guiso, L. and Jappelli, T. (2002). Private transfers, borrowing constraints and the timing of homeownership. Journal of Money, Credit and Banking, pages 315–339.
24. Huggett, M. (1993). The risk-free rate in heterogeneous-agent incomplete-insurance economies. Journal of economic Dynamics and Control, 17(5-6):953–969.
25. Hurd, M. D., Smith, J. P., and Zissimopoulos, J. M. (2011). Intervivos giving over the lifecycle. Working paper.
26. Kiyotaki, N. and Moore, J. (1997). Credit cycles. Journal of Political Economy, 105(2):211–248.
27. Lochner, L. J. and Monge-Naranjo, A. (2011). The nature of credit constraints and human capital. American Economic Review, 101(6):2487–2529.
28. Modigliani, F. (1988). The role of intergenerational transfers and life cycle saving in the accumu lation of wealth. Journal of Economic Perspectives, 2(2):15–40.
29. Mogstad, M. (2017). The human capital approach to intergenerational mobility. Journal of Political Economy, 125(6):1862–1868.
30. Nishiyama, S. (2002). Bequests, inter vivos transfers, and wealth distribution. Review of Economic Dynamics, 5(4):892–931.
31. Pasini, G., Alessie, R., and Kalwij, A. (2024). When you need it or when i die? timing of monetary transfers from parents to children. Research in Economics, 78(3):100974.
32. Pfeffer, F. T. (2018). Growing wealth gaps in education. Demography, 55(3):1033–1068.
33. Piketty, T. (2011). On the long-run evolution of inheritance: France 1820–2050. The Quarterly Journal of Economics, 126(3):1071–1131.
34. Piketty, T. and Zucman, G. (2015). Wealth and inheritance in the long run. In Handbook of Income Distribution, volume 2, pages 1303–1368. Elsevier.
35. Popov, A. (2014). Credit constraints and investment in human capital: Training evidence from transition economies. Journal of Financial Intermediation, 23(1):76–100.
36. Spiteri, J. and von Brockdor!, P. (2023). Household wealth and inheritance transfers: Evidence from the Euro Area. Journal of Family and Economic Issues, 44(3):619–633.
37. Suero, C. (2023). Gendered division of housework and childcare and women’s intention to have a second child in Spain. Genus, 79.
38. Townsend, R. M. (1994). Risk and insurance in village India. Econometrica, 62(3):539–591.
39. Zeldes, S. P. (1989). Consumption and liquidity constraints: An empirical investigation. Journal of Political Economy, 97(2):305–346.

A Data descriptives

Table A.1 : Descriptive Statistics of Inheritance Variables

Notes: Monetary values in constant 2022 SD = standard deviation. In the econometric analysis, extreme values of net wealth, age at first inheritance, and year of first inheritance are excluded to limit the influence of outliers. Source: Authors’ own construction based on the EFF (Banco de España, 2024).

Table A.2 : Inheritances by Year and Age Group

Notes: In 2002 and 2005, the survey included fewer response options related to inheritances. All percentages are computed over total weighted households per wave (Observations weighted). Monetary values in constant 2022 EUR. Source: Authors’ own construction based on EFF from Banco de España (2024).

: Descriptive statistics by survey wave (2

nstruction based on EFF from Banco de Espa˜ na (2024). Notes: All statistics are computed u

Figure A.1 : Percentage of Households that Have Received an Inheritance, by Age Groups.

Source: Authors’ own construction based on the EFF. Notes: In 2002 and 2005 the survey includes fewer response options related to inheritances.

Figure A.2 : Percentage of Households that Have Received a Gift, by Age Groups.

Source: Authors’ own construction based on the EFF.

Figure A.3 : Mean Net Wealth, by Age Groups.

Source: Authors’ own construction based on the EFF.

Figure A.4 : Median Net Wealth, by Age Groups.

Source: Authors’ own construction based on the EFF.

Figure A.5 : Evolution of the Gini Index of Net Wealth, by Age Groups and Overall.

Source: Authors’ own construction based on the EFF.

Figure A.6 : Mean and Median Age at which the Largest Inheritance is Received.

Source: Authors’ own construction based on the EFF.

Figure A.7 : Mean Age at which the Largest Inheritance is Received, by Educational Level.

Source: Authors’ own construction based on the EFF.

Figure A.8 : Mean Age at First Birth by Education Level, by year of birth.

Source: Authors’ own construction based on the EFF.

Figure A.9 : Mean Age at First Birth by Education Level, by EFF years.

Source: Authors’ own construction based on the EFF.

Figure A.10 : Percentage Inherited by each Wealth Decile over the Total Amount.

Source: Authors’ own construction based on the EFF.

Figure A.11 : Percentage Inherited by each Wealth Decile over the Total Net Wealth in each Decile.

Source: Authors’ own construction based on the EFF.

Figure A.12 : Percentage Received as Gift by each Wealth Decile over the Total Amount.

Source: Authors’ own construction based on the EFF.

Figure A.13 : Percentage Received as a Gift by each Wealth Decile over the Total Net Wealth in each Decile.

Source: Authors’ own construction based on the EFF.

B Timing Penalty within Inheritance Size Quartiles

Table B.1 : Effect of Age at Inheritance on log(Net Wealth), by Education and Inheritance Size Quartile

Notes: Dependent variable is log(Net Wealth). Quartiles (Q1–Q4) are defined over the distribution of log(inheritance value) within each education group. All regressions include household-level controls and survey-wave fixed effects. Robust standard errors in parentheses. Significance levels: *** p < 0.01, ** p < 0.05, * p < 0.1.

C Fertility and Family Planning

Fertility decisions involve substantial upfront and ongoing costs, making them highly sensi tive to liquidity and credit access. Early inheritances can ease these constraints and facilitate family formation. To quantify this effect, we estimate probit models for two outcomes: (i) the probability of having any children and (ii) the probability of having at least two children, as functions of inheritance timing and size, controlling for socio-demographic characteristics and year fixed effects.

Table C.1 : Marginal Effects on the Probability of Having Any Children

Notes: All regressions include controls for age, health status, marital status, income decile, household size, and year fixed effects. Robust standard errors clustered at the household level in parentheses. * p<0.10; ** p<0.05; *** p<0.01.

Table C.1 indicates that earlier inheritances significantly increase the likelihood of par enthood. In the full sample, inheriting one year earlier raises the probability of having chil dren by 0.44 percentage points (p<0.01). The effect is strongest among primary-educated households (0.83 pp), remains sizeable for tertiary-educated households (0.32 pp), and is smaller but still significant for those with secondary education (0.21 pp). These patterns highlight that fertility decisions are particularly sensitive to liquidity constraints among the least educated, but timing also matters for more advantaged groups.

By contrast, the size of the inheritance exerts little systematic influence. In the pooled sample, the effect is small and statistically insignificant (–0.06 pp). Among tertiary-educated households, however, larger bequests are associated with a lower probability of having chil dren , consistent with wealthier couples postponing or reducing fertility. For primary and secondary groups, the scale effect is negligible, suggesting that timing rather than magnitude drives fertility responses.

We next examine the probability of having at least two children, a decision where credit constraints are particularly binding. Spain’s fertility rate is 1.3 children per woman, wel below the desired level of around two (Delgado et al. 2008). Recent evidence shows that 44.8% of one-child mothers intend to have a second child within three years (Suero 2023), yet economic barriers—especially high childrearing costs and limited credit access—often prevent this expansion.

Table C.2 : Marginal Effects on the Probability of Having at Least Two Children

Notes: All regressions include controls for age, health status, marital status, income decile, household size, and year fixed effects. Robust standard errors clustered at the household level in parentheses. * p<0.10; ** p<0.05; *** p<0.01.

Table C.2 shows that earlier inheritances also matter for family expansion. In the ful sample, inheriting one year earlier raises the probability of having at least two children by 0.29 percentage points (p<0.01). The effect is particularly pronounced for primary educated households (0.70 pp), remains sizeable for tertiary-educated households (0.42 pp), and is smaller though still significant for those with secondary education (0.25 pp). These results reinforce that timing plays a central role in enabling households—especially the most constrained—to achieve desired family size.

By contrast, the size of inheritances exerts little systematic influence on family expansion. In the pooled sample, the effect is essentially zero. It remains negligible for primary and secondary households, but among tertiary-educated families larger bequests are associated with a lower probability of having two or more children (–0.15 pp, p<0.10). This pattern suggests that for more advantaged households, sizeable transfers may induce postponement or reduction of fertility, consistent with alternative investment or lifestyle choices crowding

out family expansion.

Taken together, the results show that early inheritances ease liquidity constraints and significantly promote both entry into parenthood and the transition to larger families. The effects are strongest among households with lower education, consistent with tighter credit constraints, but timing also matters for more advantaged groups. By contrast, the size of inheritances plays a limited role and is only significantly negative among tertiary-educated households, where larger transfers may delay or reduce fertility. These patterns underscore that the timing channel, rather than the scale of transfers, is the primary driver of fertility responses.

Table D.1 : Effect of Age at Inheritance on log(Net Wealth), by Income groups

Notes: This corresponds to the same regression explained in Equation 3. We now divide the sample by percentile groups of the household income distribution . All regressions include household-level controls and surveywave fixed effects. Robust standard errors clustered at the household level in parentheses. Significance levels: *** .

Table D.2 : Marginal Effects on the Probability of Homeownership, by Income groups

Notes: This corresponds to the same regression explained in Equation 3. We now divide the sample by percentile groups of the household income distribution. Average marginal effects from probit models. All regressions include household-level controls and survey-wave fixed effects. Robust standard errors clustered at the household level in parentheses. Significance levels: *** .

Table D.3 : Marginal Effects on the Probability of Owning other Real Estate, by Income groups

Notes: This corresponds to the same regression explained in Equation 3. We now divide the sample by percentile groups of the household income distribution. Average marginal effects from probit models. All regressions include household-level controls and survey-wave fixed effects. Robust standard errors clustered at the household level in parentheses. Significance levels: *** .

Table D.4 : Marginal Effects on the Probability of Entrepreneurship, by Income groups

Notes: This corresponds to the same regression explained in Equation 3. We now divide the sample by percentile groups of the household income distribution. Average marginal effects from probit models. All regressions include household-level controls and survey-wave fixed effects. Robust standard er rors clustered at the household level in parentheses.

Significance levels: *** .

Figure D.1 : Amount of inter vivos gifts by parental occupational status.

Notes: Parental occupation is classified by skill level following the International Standard Classification of Occupations (ISCO) and used as a proxy for socioeconomic background. The sample is restricted to households in which both parents (or parents-in-law) fall into the same skill category. The horizontal axis reports EFF survey waves rather than years of gift receipt to ensure sufficient sample size within each cell.

Source: Authors’ own construction based on the EFF.

Figure D.2 : Inter-vivos Transfers and Income Deciles (a) Probability of Receiving an Inter-vivos Gift

(b) Probability of Receiving Both an Inheritance and a Gift

(c) Predicted Log(Value of Early Gift)

(d) Predicted Probability of Receiving an Early Inter-vivos Gift

Notes: Each panel reports predicted values by household income decile, holding covariates at their means. Shaded areas correspond to 95% confidence intervals. All specifications control for demographic characteristics, education, and survey-wave fixed effects.

Source: Authors’ own construction based on the EFF.