Plenty. The basis of the poverty count is a figure Statistics Canada prefers to call the "low income cut-off," LICO for short. For many years after the war, the proportion of the population earning less than that amount had been falling. But since 1975 it has more or less stalled at around 15 per cent; in recent years -- since the last recession, that is -- it has been rising, reaching 17.8 per cent in 1995. This is setting off all sorts of alarm bells.

Should it? LICO would be a more reliable indicator of poverty if it were not a moving target. Far from a constant measure of absolute deprivation, LICO is rather a shifting index of relative inequality. Last year's cut-off -- roughly $32,000 for a family of four in a large city -- was calculated to be the point at which a family would typically have to spend 55 per cent of its income just to buy food, clothing and shelter.

Why is 55 per cent the magic number? Because it's 20 per centage points more than 35 per cent. That is, Stats Can calculates that the average family today spends about 35 per cent on the bare necessities -- then adds 20 per cent. That's how LICO was first defined, in 1959. Except at that time, the average middle class family spent about 50 per cent of its income on necessities, incomes being that much lower then than they are today. So LICO then was based on the notion that a poor family was one that would have to spend 70 per cent of its income to purchase the same basket of goods.

The required percentage has been declining ever since, in line with its middle class yardstick: the less of its income the average family spends on necessities, so, too, the definition of a poor family rises ever farther above subsistence. Part of the reason, then, that more people have been found to live below the line in recent years is that Stats Can keeps raising it. If the agency were still to use the original 70 per cent standard, there would be many fewer people defined as poor. On the other hand, had it used today's 55 per cent test back in the 1950s, almost half the population would be counted as "poor." Well, so what: maybe we don't want to define what it is to be poor, in 1996, by the standards of our parents' generation. The real trouble with LICO, however, is that it makes it almost mathematically impossible to make any headway against poverty. Suppose, for example, a benificent deity were to drop upon us the $13-billion in increased aid to the poorest families that Campaign 2000, the anti-child poverty group, would like us to spend every year. Presto: poverty is eradicated.

Except, that means average incomes rise. The proportion of middle class incomes required for food, clothing and shelter accordingly drops. So, therefore, does the proportion used to define a poor family, which means to be poor is to be a little richer than before. So a good number of those families who had just been lifted out of poverty would be defined right back into it. When you use relative measures like LICO, it is theoretically possible for everyone in society to be better off, in absolute terms, without any reduction in measured poverty.

That is why many analysts, and not only conservatives, hanker after a more absolute measure, one that would not only permit accurate comparisons from year to year but would more closely conform to what most people think of when they use the word "poverty." Prof. Christopher Sarlo has developed just such a Basic Needs Index for the Fraser Institute, using real data on the cost of a basket of necessities in different parts of the country -- not unlike the market-basket approach provinces use to set welfare rates.

Yet this, too, is a relative measure, when you think about it. Not only is the choice of how much of which kinds of goods to include in the index necessarily a subjective judgement, but what is the major cost of production in most of those goods? Wages. As incomes rise, in other words, so does even a "basic needs" poverty line. Still, because it focuses on whether poor families are being properly fed, clothed and housed, rather than whether they are keeping pace with the middle class, it is a much better gauge of poverty, and of progress in combating it, than LICO.

Last year's low-income numbers led a despairing social activist to sigh this week that Canada's announced goal of eliminating child poverty by the year 2000 was "pie in the sky." Maybe so, if poverty is defined as inequality. But a goal of eliminating want is most achievable.