Intergenerational Earnings: Tax & Education Impact

The study begins by presenting empirical evidence demonstrating significant cross-country variations in intergenerational earnings persistence. It highlights the existing literature showing relatively high persistence in the US, Britain, and Southern Europe, contrasted with lower levels in Northern Europe and Canada. This observation is supported by a meta-study of intergenerational earnings persistence across countries by Corak (2006). The paper emphasizes that understanding these differences is crucial, not just for academic curiosity, but also for its policy implications. The question of whether economic fate is predetermined or influenced by public institutions is a central theme. The potential for policy intervention is raised, particularly if the observed patterns are due to borrowing constraints preventing poor parents from optimally investing in their children's human capital. The lack of quantitative research examining the effect of cross-country policy differences on earnings persistence is noted; this paper aims to fill this gap by investigating the impact of taxation and public education expenditure.

A key finding of the study is the strong negative cross-country correlation between intergenerational earnings persistence and both tax progressivity and public expenditure on tertiary education. This correlation is the main focus of the analysis. Countries with more progressive tax systems and higher levels of public spending on tertiary education tend to have lower levels of intergenerational earnings persistence. This suggests that these policy variables are significant factors influencing earnings mobility across nations. The paper posits that this negative correlation likely stems from the effect of these policies on individual and parental incentives to invest in human capital. Higher taxes and greater public investment in education may reduce the perceived return on private investments in human capital, leading to a decrease in the strength of the link between parental resources and a child's earnings. The introduction provides context by noting several existing explanations for cross-country patterns in intergenerational earnings persistence, but emphasizes the lack of quantitative work specifically addressing the impact of cross-country policy differences. This study aims to address that gap in the literature.

The core of the study involves constructing an intergenerational life-cycle model to analyze human capital accumulation and its impact on intergenerational earnings persistence. This model explicitly incorporates several key determinants of earnings persistence: progressive taxation, the efficiency of human capital investments, public education expenditure, borrowing constraints, partially inheritable abilities, inter vivos transfers from parents to children, and idiosyncratic wage shocks. These factors are chosen based on their prominence in the existing literature. The model's structure allows for a quantitative assessment of the individual contributions of each element to overall earnings persistence. The choice to use a life-cycle model is justified by the importance of considering parental financial resources during a child's formative years, even when controlling for lifetime income. The model considers two primary factors in human capital formation: a fixed endowment (abilities inherited from parents) and investments in human capital (made by parents and the government). The theoretical underpinnings of this model build upon established works by Becker and Tomes (1979, 1986) and Solon (2004). The model acknowledges the impact of diminishing returns to human capital investment, implying the existence of an optimal investment level for each child.

The model is calibrated using data primarily from the Panel Study of Income Dynamics (PSID) from 1999-2005, focusing on employed males to align with the existing literature on father-son earnings relationships. This data provides six out of eleven data moments used for parameter estimation. The data selected covers the period 1999-2005 because data on education spending and taxes are also available for this period. The study utilizes the OECD-modified adult equivalence scale for household income adjustments. Consistent with existing empirical literature (Browning, Hansen, and Heckman, 1999), the model uses a coefficient of relative risk aversion (σ) of 2 and an inverse Frisch elasticity of labor supply (η) of 3. Other parameters such as the elasticity of substitution between consumption and labor (χ), the time discount factor (δ), and the altruism parameter (α) are estimated using additional data moments. Data on average hours worked, asset holdings, and the wage of high school graduates are also incorporated. For tax data, the study leverages the OECD tax database (2001-2005), which is well-suited for cross-country comparisons. Consumption tax data is obtained from Vertex Inc., estimating an 8.4% average consumption tax in the US in 2002. Data on college degrees and college borrowing limits are drawn from the PSID (1999-2005) and Lochner and Monge-Narajano (2008), respectively.

The calibrated model is used to decompose the contributions of different model elements to intergenerational earnings persistence. Individual investments in human capital emerge as a significant driver, accounting for 62% of the estimated intergenerational earnings elasticity in the US. Inheritable abilities/family endowments are also identified as significant. The analysis systematically examines the effect of removing each element (parental investments, inter vivos transfers, correlation of abilities, and variance of idiosyncratic shocks) while holding others constant to isolate their individual influence. Idiosyncratic wage shocks generally reduce earnings persistence due to their random and uncorrelated nature across generations, though an exception arises when only inter vivos transfers are present. Inter vivos transfers are shown to have a complex impact, with their absence limiting investment opportunities for children from rich families, potentially increasing persistence through increased human capital investment by wealthy parents, yet leading to negative intergenerational earnings elasticity when considered in isolation due to negative income effects on labor supply. The results highlight the interactions between various factors within the model, illustrating the multifaceted nature of intergenerational earnings persistence.

The study uses a calibrated model to disentangle the influence of different factors on intergenerational earnings persistence in the US. Four primary elements are identified: the inheritance of abilities from parents to children, the variance of idiosyncratic productivity shocks, inter vivos transfers from parents to children, and investments in human capital (by both parents and the government). The interaction of these elements with government investment size, returns to human capital investments, taxation, and borrowing constraints is central to the analysis. To quantify the individual contribution of each element, the model systematically shuts down each factor while holding others constant. It's important to note that human capital investments cannot be entirely removed from the model, as this would result in zero wages for everyone. Therefore, government investments are maintained at a constant level relative to average earnings while other components are individually set to zero. This approach allows for the isolation and quantification of each factor's contribution to the overall earnings persistence.

The analysis examines the effect of idiosyncratic wage shocks and inter vivos transfers on earnings persistence. Removing idiosyncratic shocks leads to an increase in intergenerational earnings elasticity, suggesting that these shocks generally reduce persistence by introducing randomness uncorrelated across generations. However, an exception occurs when only inter vivos transfers are present, in which case the introduction of shocks increases transfers, strengthening the negative correlation between parental and children's earnings. The role of inter vivos transfers is further investigated. Removing inter vivos transfers reduces intergenerational earnings elasticity. The absence of transfers limits the investment capacity of children with wealthy parents, negatively affecting persistence. Conversely, without these transfers, wealthy parents can only aid their children through increased human capital investments, thereby increasing persistence. The model highlights the complex and sometimes counterintuitive effects of inter vivos transfers, demonstrating a negative intergenerational earnings elasticity when these transfers are the only factor present, an effect attributable to negative income effects on labor supply.

The study's quantitative results reveal that individual investments in human capital account for 62% of the estimated intergenerational earnings elasticity in the US. This substantial contribution underscores the importance of individual choices and opportunities in shaping earnings outcomes across generations. The model also indicates that removing idiosyncratic shocks increases intergenerational earnings elasticity from 0.47 to 0.544, highlighting the dampening effect of random shocks. Removing inter vivos transfers lowers elasticity to 0.428, emphasizing the importance of these transfers in shaping the relationship between parental resources and children's earnings. The inherent limitations of the model are acknowledged; for instance, the absence of explicit modeling of educational service supply is noted as a potential area for future improvement. This aspect might impact the results, as the human capital production function could be influenced by changing educational demand. The results emphasize that the model's findings could be sensitive to such omitted variables.

The study uses Denmark as a prime example to investigate the impact of national policies on intergenerational earnings persistence. Denmark is chosen because it stands out among the countries analyzed (Table 1) for having the highest and most progressive taxes alongside the highest public spending on tertiary education. Importantly, Denmark also exhibits the lowest intergenerational earnings persistence. By contrasting Denmark's policies with those of other countries, particularly the US, the researchers aim to quantify the effect of differing tax and education policies on earnings mobility. The high tax rates and significant public investment in education in Denmark offer a compelling case study for examining the relationship between these policy variables and intergenerational earnings persistence. The researchers introduce Denmark's tax system into the US model to gauge its impact on the intergenerational elasticity of earnings.

Introducing Denmark's tax system into the US model reduces the intergenerational earnings elasticity by 12 percentage points, to 0.35. This represents approximately 38% of the difference in earnings persistence between the US and Denmark. The high and progressive nature of Danish taxes significantly reduces incentives for private investment in education, subsequently lowering earnings persistence. The study observes that this also leads to lower college enrollment and reduced cross-sectional inequality. Increased tax progressivity disproportionately affects high-ability and wealthy individuals, thus compressing the distribution of private human capital investment and reducing intergenerational earnings persistence. A further experiment isolates the effect of tax progressivity versus the overall tax level. By applying a tax system with the same average labor income tax rate as in the US but with the same progressivity as in Denmark, the model yields an intergenerational earnings persistence of 0.423, suggesting about 40% of the difference in earnings persistence between the benchmark economy and the economy with a Danish tax system is due to increased tax progressivity and the remaining 60% is due to higher tax levels.

Introducing Denmark's public education expenditure scheme into the model further reduces intergenerational earnings elasticity by 3.6 percentage points, to 0.434. Increased public expenditure reduces the relative incentives for parental/individual education spending, although total private education expenditure increases in absolute terms due to increased overall wealth. However, the share of private expenditure in total education expenditure falls from 53% to 42%. The combined effect of Denmark's tax and education policies significantly impacts earnings persistence, with taxation playing a more substantial quantitative role. The researchers note that while lower earnings persistence might be viewed favorably, it’s crucial to consider tradeoffs. Higher and more progressive taxation alone may hinder human capital accumulation, leading to a less prosperous society, while increased public education expenditure can have the opposite effect. The study acknowledges the complex interplay between these policies and suggests that higher taxes might be necessary to finance greater public education expenditure, noting that the introduction of Danish education spending in the model actually resulted in a net increase in tax revenue, suggesting further research on optimal policy combinations is needed.

The study acknowledges the considerable attention given in the literature to the impact of borrowing constraints on both intergenerational earnings persistence and college enrollment. This section delves into the quantitative effects of altering these constraints within the model's framework. The researchers examine the effects of tightening and relaxing the college borrowing constraint, along with relaxing the assumption that borrowing is only permitted for college attendance. Furthermore, the analysis extends to consider negative inter vivos transfers, allowing parents to pass debt onto their children. This comprehensive approach allows for a nuanced understanding of how the availability of credit and the ability to transfer debt influence the model's outcomes related to intergenerational earnings persistence. The existing literature on this topic provides the rationale for including this element in the model's design, highlighting the potential for credit market limitations to affect educational attainment and, consequently, earnings across generations. The study sets out to provide quantitative evidence concerning the magnitude of these effects.

The analysis explores the effects of adjusting the college borrowing constraint on key model outcomes. Relaxing the borrowing constraint leads to a minimal increase in college completion (14%), primarily benefiting those who would derive the most significant gains from college. Average earnings increase only slightly, and intergenerational earnings persistence shows a negligible rise of only 0.4 percentage points. Conversely, tightening the constraint also results in limited effects. College completion falls, but the impact on average earnings and intergenerational earnings persistence is modest. The explanation for these minimal impacts centers on the observation that most individuals in the model are not significantly constrained by borrowing limits when making human capital investments. The majority accumulate positive assets early to fund retirement and children's education, negating the binding effect of borrowing constraints on investment decisions. The study observes that the college borrowing constraint in the benchmark economy binds for only approximately 30% of those completing college. Even when tightened, individuals can offset the impact by increasing work hours while attending college. These observations underscore the limited role of borrowing constraints in this particular model context.

The final part of the analysis allows parents to pass on debt to their children (negative inter vivos transfers). This relaxes the previous restriction of only positive transfers. Introducing this possibility leads to lower average capital holdings and a significant decrease ($53,000) in average transfers from parent to child. College enrollment also experiences a notable drop, mainly affecting those who would gain marginally from college. Average earnings fall only slightly, and the average human capital investment in college even rises (due to lower college completion and reduced investment by those who drop out). Allowing borrowing against children's future earnings results in a minor increase in intergenerational earnings elasticity (0.468 to 0.472), relative to the scenario with identical borrowing constraints but only positive transfers. The study concludes that allowing parents to pass on debt negatively impacts children from poorer families. This comprehensive analysis of borrowing constraints demonstrates their limited quantitative impact on intergenerational earnings persistence in the US context of this model. Most individuals are not constrained by borrowing limits, and the ability to work during college further mitigates potential impacts. The possibility of negative inter vivos transfers leads to only marginal changes in the model's key outcome—intergenerational earnings elasticity.

The study concludes that taxation and public education expenditure significantly influence intergenerational earnings persistence by impacting individual investments in human capital. These policies are identified as major contributors to the cross-country patterns observed in empirical research. Specifically, taxation accounts for an average of 25% of the difference in earnings persistence between the US and 10 other countries, while public education expenditure explains 35% of the difference between the US and 7 other countries. The model shows that individual investments in human capital represent 62% of the estimated intergenerational earnings elasticity in the US. These findings highlight the substantial potential for policy interventions to affect earnings mobility. The conclusion underscores the importance of considering the interplay between these policies, acknowledging that the effects are complex and can have unintended consequences. For instance, highly progressive taxation, while reducing earnings persistence, might also decrease overall human capital accumulation and societal wealth. Therefore, a nuanced understanding of the trade-offs involved is critical when designing policies aimed at influencing intergenerational earnings mobility.

The study acknowledges limitations inherent in its model. A notable omission is the explicit modeling of the supply of educational services. The researchers suggest that the human capital production function could plausibly change in response to shifts in demand, thus potentially influencing the model's results. Furthermore, the study focuses on the impact of existing policies rather than exploring optimal policy combinations. This suggests important areas for future research: investigating the supply side of education and modeling general equilibrium effects could refine the model's predictive power. Exploring optimal policy mixes that balance the potential benefits of reduced earnings persistence (increased social mobility) with the potential negative impacts on human capital accumulation and overall economic prosperity is highlighted as a fruitful direction for further investigation. The study's findings provide a strong foundation for such future research, highlighting the significant influence of policy on intergenerational earnings persistence, but also emphasizing the need for more sophisticated models to capture the complexities of these policy interactions.