of the work force because their wage is already above the minimum wage or they are not covered2 by it. [2] Secondly, many minimum wage workers are teenagers that come from families that taken as a whole are not below, or even near, the poverty level. Thus Addison and Blackburn analyze state level minimum wage policy using panel data from families with low wage workers and then they measure the effects over a long period of time to attempt to capture the complete effects of the wage floor increase. After regressing the poverty rate on wages, state specific effects, and other fixed effects, they find the poverty/wage elasticities for the 1990's as shown in table 1.
| Population Segment | Poverty / Wage Elasticity |
| Teenagers |  |
| Young Adults |  |
| Junior High Dropouts | |
| 3 Groups Combined |  |
increase in the minimum wage will result in a 
decrease in the poverty level among these two groups. [1] However, when they studied the 1980's they found that there was no statistically significant effect on the reduction in the poverty level following a minimum wage increase. This could possibly be due to the extremely tight labor markets of the 1990's. If unemployment is very low in an economy, the demand for labor becomes very inelastic. Therefore, a increase in the minimum wage will cause a less than decrease in employment. Thus average earnings will rise and it is more plausible to see a decrease in the poverty rate. This is one explanation for why they concluded the effectiveness of minimum wage policy in the 1990's but not in the 1980's.
Bruce Bartlett estimates the overall wage/employment elasticity to be about 
[2] This means that a increase in the minimum wage results in a 
decline in the employment rate. However, only 
of workers would be affected by the minimum wage hike so the real absolute elasticity is relatively larger.3 Aside from employment losses, the minimum wage can also have several other negative impacts, including:
of the minimum wage benefits are going to families that are already making at least three times the poverty level! Consider table 2, which measures the impact of the minimum wage on UK household earnings by income decile.
| Decile | % Whose Earning Rise |
| Poorest |  |
| 2nd |  |
| 3rd |  |
| 4th |  |
| 5th |  |
| 6th |  |
| 7th |  |
| 8th |  |
| 9th |  |
| Richest |  |
of UK families in the lowest of the income distribution gain from instating a minimum wage. [5]
In the US, David Neumark and William Wascher have attempted to determine the precise effects of minimum wage policy on poverty levels. In an important article they wrote along with Mark Schweitzer[6], they ran regressions of wages, worker hours, employment, and earnings on the minimum wage and other fixed effects. They included variables to account for the lagged effects of each one of these dependent variables. Though wages initially rise following a minimum wage hike, a significant part of that increase is ``given back'' in the second year. Perhaps by employers taking advantage of inflationary effects. Employment effects are hardly significant and mixed across the distribution. They find that the coefficient on earnings is initially positive, but by the second year it is strongly negative. For example, for those people earning within ten cents of the minimum wage, a increase in the minimum wage results in a 
decrease in their earnings by the second year. A statistic that is significant at the 
level. Thus, including the lagged variables more accurately captures the real effects of a minimum wage increase. This also might be a reason for Card and Krueger's unlikely results in their one year fast food industry study.
Neumark and Wascher also attempted a slightly different type of analysis in which they try to measure the impact of the minimum age on the probability of being poor.[7] This eliminates some of the bias concerning which population segments are affected and if firms are responding by cutting back on hours or fringe benefits. Their findings are summarized in table 3.
of families that are poor in year 1 will remain poor in year 2 following a minimum wage increase. In states that did not have a minimum wage increase, 
of poor families in year 1 remain poor in year 2. The difference between the two estimates is significant at the 
level and indicates the adverse effects of minimum wage policy on poverty levels. Also, of those non-poor families in year 1 became poor in year 2 with a minimum wage increase, while 
of non-poor famililes in year 1 became poor in year 2 without a minimum wage increase. A difference that is significant at the level. Thus a minimum wage hike increases the probability that poor families remain poor by 
and increases the probability that non-poor families become poor by 
Both of these estimates are also significant at the level. However, after accounting for several controls, such as, AFDC benefits and state and year effects, an increase in the minimum wage reduces the probability that someone poor in year 1 remains poor in year 2 by 
It also increases the probability that someone comes out of poverty in year 2 by 
4
We will now turn to the study done by Card and Krueger in 1992.[4] They did a comparative analysis of the fast food industry in two adjacent areas of Pennsylvania and New Jersey. The minimum wage in New Jersey rose during the study while in Pennsylvania, the wages remained constant. They found that as a result of the increase, employment in New Jersey rose by 
full-time-equivalent employees (FTE's), while it fell in Pennsylvania by 
FTE's. Furthermore, they found that among workers in the New Jersey stores that were making above the minimum wage, there was a similiar reduction in employment (-2.04 FTE's). Those that were making the minimum wage accounted for the overall increase (+1.32 FTE's). This study has come under considerable criticism for many reasons and most consider it a good example of how not to do economic analysis. However, combined with the Addison and Blackburn study, it is worth considering if there is a way to model some of these results.