Abstract
The limited support for BI and the strong support for workfare-oriented policies (see chapter 1, Table 1) may well explain why there are no BI experiments but many, maybe thousands, welfare-to-work oriented experiments going on. BI and workfare can be seen as opposed ways to achieve more flexible labour markets. Peck and Theodore (2000, 124) assert that welfare-to-work experiments ‘seek to articulate a regulatory strategy concerned to make flexible labour markets work. “Work first” approaches, in particular, can be seen as part of a wider attempt to realign welfare provisions, incentive structures and work expectations in light of the “realities” of flexible employment; their aim is to (re)socialise welfare recipients for contingent work.’ However, as argued in the previous chapters, a BI can also be seen in the light of flexibility. Under a BI scheme a more flexible labour market may arise because of the elimination of minimum wage legislation and a less comprehensive legislation on employment conditions. By providing a BI unconditionally, the income and utility which potential workers derive from this no-work option serves as a floor. In principle at least, a substantial BI allows the possibility to deregulate the labour market, and in this way to combine the dynamics of American labour markets with the minimum income protection of European welfare states (see also chapter 2).
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
See e.g. Besley (1990) and Creedy (1996).
A similar point is made by Atkinson (1995a, 135): ‘If the decision is based solely on comparing the expected gain in earnings with the earnings foregone while training, a tax which is simply proportional would reduce both by the same percentage, and the balance in the equation is unaffected. If the tax is at rate t, we would simply have a factor (1-t) appearing on both sides. It is only to the extent that the tax has a graduated marginal rate, falling more heavily on the earnings of trained labour, that the return to training is reduced. Of course this is an over-simplified representation, and costs such as university fees may well not be tax-deductible, but the essential point is that human capital investment largely takes the form of foregone earnings, so that if these earnings would have been taxed, the cost of the investment is reduced as well as the benefits.’
Present social benefits are surrounded by all kinds of obligations (to apply for jobs, to retrain, to fill in forms every month, etc.), whereas a BI can be seen as a kind of (anonymous) gift. The sociological gift-exchange theory predicts that gifts have the tendency to elicit a counter-gift.
See e.g. Kennan (1995), Dolado et al. (1996), Greenaway (1996) and Card and Krueger (1995).
‘To understand family income, one would have to understand not only the process generating other private income sources (dividends, interest and rent) and public income sources, but also the joint decision-making process among family members who adjust their labor supply, human capital, household formation and childbearing decisions in reaction to changes in outside sources of income, as well as to changes in the earnings of other family members’ (Gottschalk 1997, 22).
See Robeyns (2000) for a more extensive gender analysis of the responses of women (specified by groups distinguished by earning capacity and labour market attachment) to the introduction of a basic income.
See the simulation results of Atkinson and Sutherland (1988) and Atkinson (1995a) using TAXMOD and of Gelauff and Graafland (1994) using the model MIMIC of the Dutch Central Planning Bureau.
For other sources, see e.g. Masters and Garfinkle (1977), Burtless and Hausman (1978), Keeley et al. (1978) and Munnell (1987). See Widerquist (2004) for a (20 pages long) bibliography, academic as well as non-academic articles, on the NIT experiments.
Two welfare reform bills issued by President Nixon embodying the idea of a NIT, the Family Assistance Plan and H.R. 1, even passed the House in 1970 and 1971 respectively, but both were ultimately rejected by the Senate. For details about the political stratagems going on behind these proposals, two books can be highly recommended: Moynihan (1973) and Lenkowski (1986).
See also the figure in Moynihan (1973, 83) which illustrated the following: ‘As male unemployment rates had gone up, so had the number of new AFDC cases. Down, down. Up, up. The correlation was among the strongest known to social science... Then with the onset of the 1960s the relationship weakened abruptly, and by 1963 vanished altogether. Or, rather, reversed itself. For the next five years the nonwhite male unemployment rate declined steadily and the number of AFDC cases rose steadily.’
‘Dependency had become a social condition beyond the apparent power of social policy to affect, save possibly at the margin. This was the heart of the Administration’s understanding of the matter. It is not a judgement that will be found in the archives. It was not even a judgement. It was simply an awareness of the limits of knowledge that gradually emerged and thereafter did not need to be dwelt upon or even acknowledged’ (ibid., 353).
The NIT-plan was not ‘predicated on the assumption that people don’t want to go to work’ (Moynihan 1973, 340).
Lenkowski (1986, 56) states that ‘... one of the bedrocks of the existing policy [was] the tradition of not providing income on the basis of need to those able to work’.
Better conditions of employment reduce the role of net wages as an incentive to elicit work effort and labour supply. Hence, lower net wages due to the higher required tax rates to finance a BI or NIT scheme will have a lower negative effect on labour supply.
The highest attrition or drop out rate, 50%, was expected for the controls, since they would only receive the fees for filling the forms and interviews. This is one of the reasons why the number of controls (632) is almost as large as the number of experimentals (725).
‘The people who drop out of the experiment may be the same as those who fail to be included in a national program (those who fail to register and those who fail to report their income or in other ways fail to maintain their right to benefits). To the extent that this is the case, the behavioral responses measured in the experiment will be a good measure of the responses that may be expected in the population as a whole. To the extent, however, that the people who drop out of the experiment differ significantly from those that remain and to the extent that they could be expected to be included in a national program, the estimates will be biased in a way that will impair the usefulness of the experiment as a guide to a national program’ (I, 117–8).
A major drawback of this decision was that families with both spouses working regularly were underrepresented.
The filing fees were forfeited if the income report form was not returned within four weeks. This filing fee was introduced ten months after the experiment started (I, 54).
J Tobin, J. Pechman and P. Mierscowski, Is a Negative Income Tax Practical?, Yale Tax Journal 77, 1967, 1–27.
Especially the in this paper much cited Vol. I and the appendices, which contain a chronology of events, and official descriptions of the rules of operation for the programme, the definition of income, the definition of the family unit, the filing periods and procedures and the review board.
The researchers involved in the experiment never agreed on, or set down in advance, a summary of what they felt was the most likely outcome for labour supply. ‘In retrospect, this is unfortunate. Any attempt to do so now is bound to reflect, to some extent, our present knowledge of the results and thus understate the degree to which we have been surprised’ (II, 14).
‘Consider first the male household head with a steady job involving hard work and long hours. If he knew that negative tax payments were permanent, he might instead take a job with lighter work and more normal hours. Yet for a period of three years, such a shift might seem too risky. At the end of the experiment, he would need the higher earned income but might be unable to get his old job back. For the steadily employed male head, the probability is that an experiment of limited duration will have smaller effects on labor supply than will a permanent program... Wives, teenagers, and other adults in the household are likely to be in and out of the labor force as family circumstances change. To the extent that periods of withdrawal from the labor force are planned in advance, a temporary experiment encourages the concentration of such periods during the experimental years, when the costs of not working are lower than normal’ (II, 10).
This section is written in close collaboration with Paul de Beer.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer Science+Business Media New York
About this chapter
Cite this chapter
Groot, L. (2004). Why Launch a Basic Income Experiment?. In: Basic Income, Unemployment and Compensatory Justice. Springer, Boston, MA. https://doi.org/10.1007/978-1-4020-2726-0_6
Download citation
DOI: https://doi.org/10.1007/978-1-4020-2726-0_6
Publisher Name: Springer, Boston, MA
Print ISBN: 978-1-4419-5251-6
Online ISBN: 978-1-4020-2726-0
eBook Packages: Springer Book Archive