Active Labour Market Policies (ALMPs), which provide training and subsidised employment to the unemployed, are an important part of Ireland’s welfare state. While a good deal of existing research is concerned with the effect of these policies on employment chances and on wage rates, none addresses the connection between poverty and ALMPs. Do these policies have an effect on poverty? That is, first, to what extent do these policies serve the low-income population, as a consequence of and in addition to their focus on those in precarious labour market situations? Second, to what extent do these policies function to lift people out of poverty in the medium term? To address these issues we use longitudinal data from the Living in Ireland Survey (1994–2001) and examine how the respondents’ situation in one year predicts participation in employment and training schemes in the next year, and then how participation in these schemes affects poverty status in the following year. Participants on both sorts of schemes are much poorer than the population average, and those on employment schemes (but not training schemes) are even poorer than one would expect given their observed characteristics. Employment schemes and training schemes serve different purposes and different populations. A conventional logistic regression analysis seems to suggest that employment schemes (but not training schemes) positively increase the risk of poverty in the following year. This finding is not considered reliable, but rather it reflects the selection processes whereby those on employment schemes are in particularly vulnerable situations, in respects that are not picked up in the data set. A more rigorous analysis, using propensity score matching, reveals that employment schemes are neutral on poverty risk. Training schemes have a weak but insignificant protective effect. Considering the risk of poverty approximately one year after participation begins, employment schemes (and to a lesser extent, training schemes) do not provide a mechanism for immediately exiting poverty. We add the caveat that it may be desirable to consider outcomes two or more years into the future, were data available, and that other outcome measures of quality of life should also be taken into account. Ultimately, with regard to both labour market and poverty outcomes, we find no evidence that participants of training schemes or employment schemes have either raised their employment chances or reduced their risk of poverty in the year following their participation.
