A few figures (and thoughts) on the Portuguese low income maintenance program, the "Social Integration Income''. Firstly, it is very low, by any standards. The reference individual income amounts to 187 euros a month (the minimum wage is equal to 450 euros). As usual, the transfer is equal to the difference between the reference value and the total household income. Appropriate equivalence scales are applied. The average individual transfer, as of 2008 (last available detailed data, can be found here in Portuguese) was as low as 86,74 euros a month. The total budget of the program was roughly 425 million euros in 2008 and 507 million in 2009. For 2008, again, this represents around 2% of the total Social Security expenditure. The S&GP defines a ceiling for the total spending with this (arguably modest) program. The ceiling aims at putting the program back into its 2007 expenditure by 2013 (370 million euros).
True, the total expenditure with the program has been increasing since its introduction in the late 90's. There are (at least) two reasons for this. The first is just the natural evolution of a transfer program. Initially, potential beneficiaries are not aware of their entitlements and it takes time for the program to reach its steady state. The second one is obvious: the pervasive crisis that has hit the country and the ever-increasing unemployment levels.
Everyone knows that there is fraud in transfer programs. That is just the natural consequence of running them. These programs are characterized by two types of errors: some people who get the transfer should not actually get it, while some who are in a situation of deprivation and hence should belong to the population of beneficiaries do not. Unfortunately, each error is the mirror of the other, and it is very hard to fight fraud without leaving deserving people out of the program. The Portuguese program already has several mechanisms to minimize the first type of error, and this can no doubt be improved. While I do wish the savings that the government aims at can be attained just by doing so, i.e., targeting the first type of error, I very much doubt that they can be achieved without worsening extreme poverty situations, particularly taking into account the hard times that are to come.