Optimal redistributive policies by publicly provided inputs and income taxation

• Published date: 01 June 2022
• Author: Thomas Bassetti,Luciano Greco
• Date: 01 June 2022
• DOI: http://doi.org/10.1111/jpet.12561

Received: 14 May 2020

|

Accepted: 30 November 2021

DOI: 10.1111/jpet.12561

ORIGINAL ARTICLE

Optimal redistributive policies by publicly

provided inputs and income taxation

Thomas Bassetti

1

|Luciano Greco

2

1

Department of Economics and

Management, University of Padua,

Padova, Italy

2

Department of Economics and

Management and CRIEP, University of

Padua, Padova, Italy

Correspondence

Department of Economics and

Management and CRIEP, University of

Padua, via del Santo 33, 35123 Padova,

Italy.

Email: luciano.greco@unipd.it

Funding information

Università degli Studi di Padova,

Grant/Award Numbers: CRIEP research

funds, Progetto SID 2018

(GREC_SID18_01)

Abstract

Governments redistribute income through taxes,

transfers, and public services. Using three key statis-

tics, we characterize the conditions under which non-

linear income taxation is optimally combined with a

publicly provided input in economies where individual

wages are driven by households' exogenous abilities

and unobservable input investments. A universal

scheme in which all households opt for a large and

uniform level of publicly provided input optimally

compounds with nonlinear income taxation if input

and ability are substitutes in households' earnings.

Calibrating our model with US data, we find that this

scheme represents the optimal policy regime for a wide

spectrum of plausible parameter values.

1|INTRODUCTION

A large part of developed and emerging countries' public social expenditures finances the

provision of in‐kind benefits that affect households' productivity (Currie & Gahvari, 2008).

1

Measures of inequality show that publicly provided social services involve a substantial re-

distribution between households in addition to the distributive impact of tax‐and‐transfer

schemes (Aaberge et al., 2017; OECD, 2011). Together with strained public finances, these

considerations motivate investigations into the appropriate design of redistributive policies that

J Public Econ Theory. 2022;24:504–528.wileyonlinelibrary.com/journal/jpet504

|

© 2021 Wiley Periodicals LLC

1

In 2015, the OECD countries' average public social spending was 20.7% of GDP, where 12.3% of GDP‐financed cash

benefits, and the remaining 8.4% of GDP went to finance in‐kind benefits (http://www.oecd.org/social/expenditure.

htm). OECD countries are heterogeneous with respect to both social expenditures and the fraction of in‐kind benefits.

As for the impact of public (and private) spending on agents' productivity, we observe that childcare, education, and

healthcare affect human capital accumulation, and childcare and elderly care also influence households' (market)

productive capacity by reducing the need for informal work in the family.

mix income taxation, cash social benefits, and in‐kind transfers (OECD, 2015). This paper

studies a policy mix characterized by the public provision of a productive input and the non-

linear taxation of labor income.

In our paper, wages (or productivity) are affected by both innate abilities and human capital

accumulation (see Stantcheva, 2017), where the latter can be done as public or private

investment.

2

Hence, the optimal redistribution policy crucially depends on the Hicksian

coefficient of complementarity between the two factors in the wage function. In this respect, if

input fosters the household's capacity to exploit exogenous productive ability (e.g., childcare,

higher education), then the two factors are complements in the wage function (i.e., the

Hicksian coefficient of complementarity is above 1). Conversely, if the input compensates for a

lack of exogenous ability (e.g., if school attendance widens the household's social network), the

two factors are substitutes in the wage function (i.e., the Hicksian coefficient of com-

plementarity is below 1).

Compared to a pure taxation regime, the public provision of private goods affects how

marginal tax rates distort households' choices and, thus, the optimal tax schedule (Blomquist

et al., 2010).

3

In particular, it may curb the efficiency cost of tax‐and‐transfer redistribution

through two channels: tax distortions are tamed by forcing households to use more of the

publicly provided good (e.g., Boadway & Marchand, 1995; Cremer & Gahvari, 1997) and in‐

kind transfers improve the redistribution capacity of public policies by self‐targeting (e.g.,

Blomquist & Christiansen, 1995).

The design of the public provision mechanism is also importan t for redistribu tion effi-

ciency (Blomquist & Christiansen, 1998a,1998b;Greco,2011). Considering that individual

private transactions in goods and inputs are not always perfectly observable, real‐world

public‐provision programs can be implemented in two conceptually different ways.

4

First, the

government can support households' expenditures to reach a given quantity or quality

of a specific good or service using conditional transfers—forexample,vouchersortax

allowances—leaving households free to top up the basic public provision (topping‐up

scheme). Second, the government may provide a public option as an alternative to market

services (opting‐in scheme). Under such a scheme, households are free to choose the public

option or alternative private services, but private supplementing is legally or technologically

unfeasible once they opt for the former. For example, a child can attend only one school.

Mixed schemes, where topping‐up and opting‐in mechanisms coexist, have been observed in

various countries and times. For example, in several countries, the government provides free

education through public schools, affording, at the same time, tax allowances to households

that send their children to private schools.

We characterize, both theoretically and through a numerical simulation, optimal redis-

tributive policies considering the design of public provision of a productive input in a model of

optimal nonlinear taxation with a continuum of types (i.e., innate abilities) of individuals.

2

Wage endogeneity may also arise from the equilibrium effects of tax changes on the labor market (see e.g., Ales et al.,

2015; Stiglitz, 1982). In this paper, we abstract from these effects.

3

A thorough survey of the theoretical literature on publicly provided private goods is beyond the scope of this paper.

Currie and Gahvari (2008) review this literature and contrast it to empirical evidence for the United States and other

developed and emerging countries.

4

Our analysis does not consider the case of nonlinear taxes and subsidies on goods and inputs because of the

nonobservability of individual transactions. Assuming that the individual transactions of the productive input are

observable, Bovenberg and Jacobs (2005) and Jacobs and Bovenberg (2011) show that input subsidies mitigate

nonlinear income‐tax distortions.

BASSETTI AND GRECO

|

505