Plank 1 solves inequality

The discussion on the merits of adopting unconditional basic income programs is a long-dated debate that is gaining momentum at present. Several countries in Europe are exploring the possibility and some have already passed basic income proposals; in the United States, there are calls to reform the entire welfare system (Steuerle, 2012; Murray, 2006; Alexander, 2013; among others) and there is an ongoing debate on whether to expand the Earned Income Tax Credit (EITC) to a broader audience.1 Under an unconditional basic income program, all citizens are assured a minimum income and there are no means tests or work requirements on behalf of the recipients, making it an effective tool to fight poverty without the administration costs of the current web of welfare programs. Naturally, a universal basic income can be instrumented in several ways and, following Friedman (1962), a Negative Income Tax (NIT) stands out because of its simplicity and progressive structure, thus becoming a valid candidate to replace the actual income tax and welfare system. A NIT does not distort the price system, as other income policies do, like the minimum wages policies or price controls, and it is able to eradicate the so-called welfare cliffs.2 It is no coincidence that a NIT is considered “one of the fundamental ideas of modern analysis of welfare programs” (Moffit, 2003). However, the effects that a NIT has on the labor supply or the increase on tax rates necessary to fund this minimum income is not clear yet. These are not minor questions and they need to be analyzed in the light of a general equilibrium model. In this paper, I will make a case for a NIT and carry out the first, to the best of my knowledge, quantitative analysis of the tax in a general equilibrium setting. Specifically, I will answer the following questions: What are the macroeconomic effects of replacing the actual U.S. income tax and the welfare system with a NIT on income and earnings, labor supply, tax rates, savings, and welfare? Should we pick a NIT? The present state of affairs makes the NIT a relevant and timely tax reform. On the one hand, the federal income tax has become increasingly complex and distortionary. Its considerable number of tax credits, deductions and overlapping provisions creates differences in the amounts paid by households earning the same level of income which lessens its progressivity and creates a fair share of distortions at all income levels. On the other hand, the welfare system makes it harder for low- and middle-income households who see their marginal tax rates surge at particular income thresholds, as many of the welfare programs in place vanish as their income increase. According to the Congressional Budget Office (Congressional Budget Office, 2012), 51% of taxpayers below 450% of the Federal Poverty Line face marginal rates higher than 30%, and 8% of them are taxed with marginal rates higher than 50%.3 In contrast, only 5% of the high-income households are taxed at such high rates.4 In its simplest version, a NIT combines a constant marginal tax rate and fixed lump-sum transfer to all households. It works as follows. At the beginning of the fiscal year, all households receive a transfer from the government, say $2000, and all income made is taxed at a constant rate, say 20%. If a household has a yearly income of $50,000, then its total tax payments for the period will be $8000 ($50,000 20% – $2000). This means that households earning less than $10,000 ($2000C0.2) pay no taxes and receive a positive net transfer (negative tax). All households have a guaranteed minimum income, and as income increases, the effect of the transfer declines. To demonstrate a NIT's suitability, I study a life-cycle economy, in which agents are ex-ante homogeneous but grow different over time as a result of life uncertainty, together with age-independent and idiosyncratic productivity shocks. At any point in time, the resulting heterogeneity is characterized by the agents' shock history, their level of asset accumulation, and their age. Welfare payments are modeled as a non-linear function of income to capture the progressive structure of the U.S. welfare system. In addition, there is a social security system; once retired, agents receive benefits in the form of lumpsum payments.5 I calibrate this model to match several features of the U.S. economy, reproducing the actual distributions of labor earnings, wealth, federal income tax liabilities, and transfers. I focus on an equilibrium with transitional dynamics for an open economy and find the level of transfers and marginal tax rate such that the NIT reform is revenue-neutral and maximizes ex-ante welfare (i.e., expected utility prior to birth and revelation of agents’ types) for an agent born at the time of the reform, taking into account the transitional path to the new steady state. My findings can be summarized as follows. First and foremost, the NIT produces important welfare gains. A NIT with a marginal tax rate of 22% and a transfer of 11% of per-capita GDP – roughly, $5800 – implies a welfare gain equivalent to 2.1% annual increase in consumption. Low-ability agents in the bottom quintile of the productivity distribution benefit the most, with welfare gains that range up to 25%, but there are losers under the NIT: those in the upper level of the productivity distribution. Under an optimal NIT, the degree of redistribution is not trivial. Second, the size of the transfers plays an important role in the results. Indeed, a proportional tax – that is, a NIT with no transfers (a “non-negative income tax”) – has a welfare loss of 4.1% relative to the current tax system. This result is unsurprising; redistribution is an important feature of the U.S. income tax and welfare system. The elimination of transfers benefits only the most productive agents who, as a result, face a lower marginal tax rate. The way transfers are designed is also important. A competing scenario where the welfare system is kept unchanged and a flat tax is enacted (a “non-negative income tax” with a fixed deduction instead of a transfer) likewise underperforms the NIT. In particular, the optimal flat tax (characterized by a marginal tax rate of 15% and a fixed deduction of 29% of per-capita GDP, roughly $15,400) produces a 0.4% welfare gain. Third, there is a negative relationship between the magnitude of the NIT's transfer and per-capita GDP, which decreases by 9% under the optimal NIT. As leisure is a normal good, this drop in per-capita GDP does not automatically translate into a fall in welfare. Quite on the contrary, as the transfer insures agents against periods of low productivity, they are now able to work more hours when they are more productive, increasing their consumption of leisure and reducing their need to save in order to buffer the impact of negative income shocks. The savings rate drops 13% – implying a 22% reduction in the capital stock – and the composition of the labor force changes, where agents with high-productivity shocks are now working more hours than low-productivity agents. The immediate consequence is that the Gini coefficient for labor earnings increases to 0.57 from 0.55, but not consumption inequality, which is reduced due to the effect of the lump-sum transfer from the NIT. Finally, the optimal NIT has a strong support in the population. However, whether the tax may be transformed into a law is an open question. For the sake of understanding how sensitive to particular voting rules the NIT design is, I search for a NIT with a level of transfers and a tax rate that has the minimum required support in the population to be enacted (i.e. only 50% of the households favor the new tax), and I call it a “Popular NIT.” Even though I am not considering a political economy model, this “Popular NIT” shows that in order to gain support from wealthy capitalists and high-income earners, its rate and transfer have to be lower than the optimal NIT (19% tax rate and a 9% transfer). Interestingly, this NIT produces a considerable welfare gain of 1.7% and still outperforms a flat tax reform. The “Popular NIT” has a stronger role in the economy's efficiency, giving the most productive agents a reduction in tax rates. All productivity types but the second most productive are better off, and there are still considerable welfare gains for less productive agents. Capital and GDP fall by less than in the optimal NIT – 6% and 2% respectively – and the labor supply, measured in efficiency units, has no changes despite the 3% plunge in hours worked.