The combination of higher total real earnings but fewer people employed means that poverty and ultrapoverty could have risen or fallen, depending on where the winners and losers are located in the income distribution. Although the LFS data are not adequate for a full poverty analysis, I can compare the estimated elasticities of employment derived here to some critical values I estimated from earlier research, using the 1993 PSLSD data (Hertz 2002). This comparison suggests that the cumulative two-year changes in wages and employment brought about by the new regulations should, on balance, have reduced poverty somewhat, particularly for the households of male domestic workers, although this conclusion is necessarily tentative.