# The case for NIT+FT in Europe. An empirical optimal taxation exercise

Abstract: We present an exercise in empirical optimal taxation applied to a Negative Tax with Flat Tax reform for a sample of eight European countries: Austria, Belgium, France, Germany, Ireland, Italy, Luxembourg and the United Kingdom. Two popular approaches to empirical optimal taxation are the structural analytical approach and the non-tructural “sufficient statistics” approach. This paper presents a different approach that combines structural microeconometric modelling, behavioural microsimulation and numerical optimization. For each country, we estimate a microeconometric model of labour supply for both couples and singles. A procedure that simulates the households' choices under given tax-transfer rules is then embedded in a constrained optimization program in order to identify optimal rules under the public budget constraint. The optimality criterion is the class of Kolm's social welfare function. The tax-transfer rules considered as candidates are members of a class that includes as special cases various versions of the Negative Income Tax: Conditional (means-tested) Basic Income, Unconditional Basic Income, In-Work Benefits and General Negative Income Tax, combined with a Flat Tax above the exemption level. The analysis in most cases show that: the General Negative Income Tax strictly dominates the other rules, including the current ones; the Unconditional Basic Income policy is better than the Conditional Basic Income policy; Conditional Basic Income policy may lead to a significant reduction in labour supply and poverty-trap effects; In-Work-Benefit policy is strictly dominated by the General Negative Income Tax and by the Unconditional Basic Income. We also exemplify the possibility of identifying the mapping between the contry-specific “primitives” (social preferences, productivity, public budget constraint, labour supply elasticity and Gini coefficient) and the optimal tax-transfer rules.

## Summary (2 min read)

### 2. The alternative policies

- Note that GNIT is by definition more general than the other NIT special cases.
- Yet it is important to define the optimal design of the special cases: although necessarily not superior to GNIT according to the Social Welfare criterion, they might be more attractive than GNIT according to other dimensions that are not taken into account by the Social Welfare function.

### 4.1 Household preferences and choices

- Households can choose within an opportunity set containing jobs or activities characterized by hours of work h, sector of market job s (wage employment or self-employment) and other characteristics (observed by the household but not by us).
- The authors define h as a vector with one element for the singles and two elements for the couples,.

### D

- For couples, i D contains two analogous sets of variables, one for each partner.
- The parameter estimates of the behavioral models for singles and couples for six countries (Belgium, France, Ireland, Italy.

### 4.2 Data

- The datasets used in the analysis are the EUROMOD input data based on the European Union Statistics on Income and Living Conditions (EU-SILC) for the year 2010.
- The input data provide all required information on demographic characteristics and human capital, employment and wages of household members, as well as information about various sources of non-labour income.
- The authors apply common sample selection criteria for all countries under study by selecting individuals in the age range 18-65 who are not retired or disabled.
- Then EUROMOD 4 provides calculations of household-level tax and transfer liabilities given the household characteristics and gross incomes according the existing tax and transfer rules.
- The target population consists of all private households throughout the national territory in every country.

### SMRS e e

- The simulation results presented in Section 5 are based on 0.05.
- Absolute indexes are less popular than relative indexes (e.g. Gini's or Atkinsos's), although there is no strict logical or economical motivation for preferring one rather than the other.
- 7 Blundell and Shephard (2012) adopt a social welfare index which turns out to be very close to Kolm's.
- Their main motivation for their index seems to be the computational convenience, since it handles negative numbers (random utility levels, in their case).
- The authors motivation in choosing Kolm's index is analogous.

### 4.5. Identifying the optimal policies

- It is important to keep in mind that the simulated policies differ from the current policies with respect to many dimensions.
- First, the authors simulate policies with a FT, while all the countries included in the present exercise adopt increasing marginal tax rates.
- In general, while the current systems might be somehow close to CBI or IWB or other versions of NIT-like mechanism, they are much more complicated.
- The comparison of the reforms to the current system is informative upon the effects of the reformed budget sets, including the effects of the universal and permanent extention to the whole population.
- It is not directly informative upon dimensionssuch as the administration costswhich are not represented in their microeconometric model.

### 5. Results

- The Tablesone for each countryshow, for each of the policies considered, the optimal tax-transfer parameters, the average individual labour supply, the household poverty rate, the percentage of household winners with respect to the current system and the change in the money-metric social welfare as percentage of the average household available income.
- Labour supply is measured by average annual hours of work (including the zero hours of the non-employed).
- The poverty rate is the percentage of households with available equivalized income below 60% of the median equivalized income.
- Let W0 and WP respectively the Social Welfare levels attained under the current regime and under a certain policy.
- Note that they are monetary measures .

Did you find this useful? Give us your feedback

...read more

##### Citations

14 citations

9 citations

6 citations

1 citations

### Additional excerpts

...For instance, Islam & Colombino (2018), Aaberge and Colombino (2013) and many others adopt a procedure that consists of using a common utility function as an argument of the social welfare function following Deaton and Muelbauer (1980) approach18....

[...]

...Islam & Colombino (2018)....

[...]

### Cites background or methods from "The case for NIT+FT in Europe. An e..."

...Islam and Colombino (2018) identify optimal tax-transfer rules within the NIT+FT class in eight European countries....

[...]

...where z denotes taxable income, h(z) and H(z) are the density and distribution functions, ( ) z is a social weight assigned to people with income greater than or equal to z and z is the elasticity of z with respect to (1-T’(z)). However, the optimal z and its distribution (and possibly also z ) depend on the optimal tax function (.) T . Therefore, in order to be able to compute the optimal taxes we must specify how z, H(z) and h(z) depend on (.) T . In other words, we must go back to Mirrlees (1971) as in Saez (2001) and Brewer et al....

[...]

...Sections 3.1, 3.3 and 3.4 hereafter follow the presentation provided in Islam and Colombino (2018)....

[...]

...This type of analysis is just sketched and exemplified in Islam and Colombino (2018). A full analysis on a large sample of countries is left for future work....

[...]

...Islam and Colombino (2018) develop a systematic procedure that combines microeconometrics, microsimulation and numerical optimization in order to identify optimal rules within the Negative Income Tax + Flat Tax class....

[...]

##### References

7,019 citations

### "The case for NIT+FT in Europe. An e..." refers background or methods in this paper

...Also as a response to the these problems, the so-called Negative Income Tax (NIT) was proposed by Friedman (1962) and supported by many authors (e.g. Tobin et al. 1967).3 The typical version of NIT is illustrated in Figure 2....

[...]

...…for other possible problems: high burocratic costs, “welfare stigma” effects and take-up costs leading to low take-up rates, incentives to underreporting of income, errors in applying eligibility criteria and litigation costs (e.g. Friedman 1962, Friedman and Friedman 1980, Atkinson 2015)....

[...]

...The original NIT proposal made by Friedman (1962) was formulated according to the second interpretation, with t1 typically larger than t2, (convex profile of the tax-transfer rule as in Figure 2)....

[...]

^{1}

3,945 citations

### "The case for NIT+FT in Europe. An e..." refers background or methods in this paper

...At least the first simulations based on empirical calibr tions of Mirrlees’ (1971) model suggested as optimal a rule very close to a NIT+FT....

[...]

...…Optimal Taxation (TOT) approach that consists of computing optimal policies using theoretical formulas with imputed or calibrated parameters, as many authors have done, e.g. using the theoretical results of Mirrlees (1971) or the more recent ones such as Diamond (1998) and Saez (2001, 2002)....

[...]

...First, Mirrlees (1971) and Saez (2001) only cover interior solutions on the part of the agents and therefore only intensiv labour supply responses are considered, while te empirical labour supply literature reveals the importance – when not the predominance – of the extensiv responses....

[...]

...This procedure is more flexible than the empirical applications of the TOT provided for example by Mirrlees (1971) or Saez (2001, 2002)....

[...]

3,864 citations

### "The case for NIT+FT in Europe. An e..." refers background in this paper

...We follow here a tradition that defines the comparability problem – as far as the empirical research is concerned – as in Deaton and Muellbauer (1980), where the adoption of the CMU approach or of the common utility are legitimate solutions....

[...]

3,050 citations

### "The case for NIT+FT in Europe. An e..." refers methods in this paper

...We choose Kolm (1976) Social Welfare index, which can be defined as:...

[...]

...We choose Kolm ( 1976) Social Welfare index, which can be defined as: ( ){ }i i exp1 ln N k W k µ µ µ − − = − ∑ (8) where 1 is an index of Efficiencyi iN µ µ= ∑ , ( ){ }i i exp1 ln Kolm Inequality Index, N k k µ µ − − = ∑ Inequality Aversion parameter,k = µi = comparable money-metric…...

[...]

2,842 citations

##### Related Papers (5)

[...]