Section 4
Who fall between the lines?

View the most recent version.

Archived Content

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.

Archived

This page has been archived on the Web.

Since aggregated indexes under different lines may behave differently in the short-run, it is interesting to see how these lines compare to each other in terms of their capabilities to identify low income individuals. For example, to what extent is an individual identified as low income by one line also identified by other lines and to what extent an individual who is above one line is also above the other lines?¹ We shall focus on the 2000 to 2007 period in this and the next sections as these are the years for which the Market Basket Measure (MBM) line is also available.

The result is contained in Table 4. The left side of the table shows that if, according to one line, an individual is above low income, what is the probability he or she would also be above low income under other lines. The right side of the table indicates, on the other hand, if an individual is identified as in low income by one line, what is the probably he or she is also identified as in low income by other lines. The top portion of Table 4 (left and right sides) shows the interaction of Low Income Cut-Offs (LICO) with variable Low Income Measure (LIM), fixed LIM and MBM. It indicates that if an individual was above LICO, he or she was much likely to be above the other lines in the 2000 to 2007 period. If an individual was not captured by LICO, the probabilities this individual would be captured by the two LIM lines varied between 1 and 2%, while the probability he or she would be captured by MBM was from 2 to 4%. On the other hand, if an individual was captured by LICO, he or she still had a relatively high probability of being above the other lines. For example, an individual being identified by LICO as in low income in 2007 had a 29% chance to be above fixed LIM.

Table 4 Percentages captured and not captured by low income lines (2000 to 2007)

The situation for variable LIM is similar. An individual who was identified as being above the variable LIM line was very likely to be above the LICO and MBM lines, and was essentially certain to be above fixed LIM. But if an individual was captured by the variable LIM line, he or she still had a strong probability of not being captured by other lines. For example, for those who had been captured by variable LIM, there was a 12% to 27% chance they would be not captured by LICO between 2000 and 2007.

The third portion of the table shows that, when an individual was identified as above low income under fixed LIM, his or her chance to fall in low income under the other three lines varied from 4 to 5% under MBM, 2 to 4% under variable LIM, and 3 to 4% under LICO. On the other hand, if an individual was identified as being in low income by fixed LIM, the individual was essentially certain to be captured by variable LIM and MBM, although there was still a chance (6 to 8%) for him or her to fall under the LICO line.

Finally, with MBM, when an individual was identified as being above low income, it was almost certain that he or she would also be above the other lines, while if the individual was captured under MBM, there was a strong possibility for the other lines to fail to capture this individual. For example, an individual who was identified by MBM as in low income in 2000 had a 68% chance to be in low income under fixed LIM, and hence, his or her chance to be above the fixed LIM line would be 32%.
 
Overall, Table 4 suggests the use of multiple low income lines can be helpful in identifying individuals who would have been miss-captured by a single line. The 1992 LIM line is an exception: in the absence of this line, all of the would-be low income individuals could have been captured by variable LIM or MBM. This is because the 1992 LIM had been fixed for eight years by 2000. When it was re-based to 2002, the re-based LIM was able to capture at least 5% of individuals who could have been missed by the MBM line (see the percentages in the parentheses in Table 4). Likewise, at least 12% of individuals would be miss-captured by LICO in the absence of the 2002 fixed LIM. This seems to suggest that, a fixed LIM line needs to be re-based periodically. Otherwise, its usefulness can be compromised.

---

Note

1. The equivalent questions are: to what extend an individual who is captured by one line is not captured by the other lines and, to what extent an individual who is above one line would fall under the others. But due to sample size restriction, these two questions cannot be examined over several disadvantaged groups of individuals.