Friday, November 25, 2011

No Alternative to Budget Cuts?

It is often argued that there is simply "no alternative" to the destruction of social safety nets implicit in proposals to cut government expenditure, e.g. the Irish government's consideration of cuts to child benefit, a reduction in unemployment benefit and the introduction of medical card fees etc., as ways to achieve the required €3.8bn "adjustment" in the forthcoming budget. This line was repeated by Stephen Collins in Saturday's Irish Times. However, the idea that there is "no alternative" to a government decision could never be accurate. The "no alternatives" myth is nothing more than a politically convenient way of saying we don't want to consider the alternatives.

Thankfully, various groups outside of government have put together detailed proposals on alternative ways of saving money in the budget. Michael Taft of TASC has outlined potential "alternative" savings of around €6bn.

It's worth noting that some of the proposals here - e.g. removal of property tax reliefs - present the relatively rare opportunity for government, in imposing taxes, of a win-win on both equity and efficiency grounds. Removal of these reliefs would also represent a reversal of the much criticized policies of the previous government. Rather than subsidizing asset accumulation, taxing wealth (in the form of a property tax or capital gains tax) makes sense as both a progressive tax measure and one which avoids the disincentivization of work entailed in higher income taxes.

On income tax, the proposal to increase the top rate to 48% is still some way below the "optimum" top rate of tax (to be applied only to the top 1% of income earners) of over 70%, as calculated in this new paper in the Journal of Economic Perspectives, and discussed here (Paul Krugman's NYTblog) and here (Kevin Drum of Mother Jones).

The calculation is based on US data. For Ireland we might consider that the "behavioural elasticity", as Krugman calls it - i.e. the reduced work effort in response to higher taxes - is likely to be higher than for the US, given the historically high international mobility of Irish workers and the infamous culture of tax exiles among Ireland's elite. I'm not sure if studies on this elasticity exist for Ireland. But there is another way of looking at this. If we assume that the current top rate of income tax in Ireland is the optimum rate, what degree of "behavioural elasticity" would this imply? Ireland's top marginal rate of income tax is officially 41%. However, if we include the top rate of the Universal Social Charge (USC) at 7% and PRSI at 4% we have a top marginal rate of 52%. This implies a "behavioural elasticity" of about 0.62.* This is still higher than even the most conservative upper bound estimates for the US - around 0.57. Also, as Diamond and Saez note, most of the behavioural response observed in studies to date consists of tax evasion or avoidance behaviour, and not the kind of changes in real economic behaviour that opponents of higher income taxes claim to be concerned about (e.g. labour supply, business creation and savings decisions).

So while increases to the top rate of tax will be dismissed by this government on pragmatic grounds - that such increases would be self-defeating - such claims may not stand up to close scrutiny. Imposing a higher rate of tax on the top 1% of earners in society - and ensuring they actually pay their tax by closing loopholes, removing tax reliefs, and clamping down on tax exiles and tax evasion - could offer a more socially equitable and sustainable means of closing the gap between public revenue and expenditure.

Certainly the notion that we have "no alternative" to the proposed cuts, appears to have no basis in the available facts and figures. Instead it is revealing of an ideologically driven budgetary process.

*The calculation is based on the formula used in the paper by Diamond and Saez cited above: T = 1/(1 + ae), where T is the optimal top rate of tax, a is the Pareto parameter that describes the distribution of incomes at the top, and e is the behavioral elasticity. I assume that the distribution of incomes at the top in Ireland is roughly equivalent to that in the US and so I use the same Pareto parameter as used in the Diamond and Saez paper, i.e. 1.5.

No comments:

Post a Comment