December 12, 2011 – On Saturday, MLI Director of Research Jason Clemens discussed how income inequality “isn’t as simple as it’s portrayed” in the Vancouver Sun. His column is below:
By Jason Clemens, Vancouver Sun, December 10, 2011
Occupy Wall Street and growing unease over unemployment have heightened interest in the near perpetual issue of inequality. Far too many analyses of inequality, however, treat it as a simple, straightforward issue when the opposite is true. Inequality is terribly complicated and simplistic diagnoses risk worsening the problem.
Take the measurement of inequality, which normally relies on both official income statistics as well as surveys. The simple comparisons normally employed ignore the influence of unreported income. Numerous studies have concluded that certain types of income are consistently underreported in Canada. For instance, self-employment income is usually under-reported by 12 to 24 per cent. Similarly, transfers from government for employment insurance have been demonstrated to be under-reported by 20 per cent and social assistance by 40 per cent.
Government sources of income are largely concentrated at the lower end of earnings. Thus, at least some of the income inequality may in fact be due to unreported income.
Further, consider this problem of under-reporting income and its effect on inequality statistics in an international context. Many countries, including both industrialized countries like Greece and most developing countries, have much larger informal or underground economies than developed countries like Canada, which affects the calculation of earnings for lower-income households.
A different question is whether we should be concerned with income or consumption. If our primary interest is ensuring that households in the bottom 20 per cent have adequate resources to secure basic necessities then consumption, not income, should be the focus.
Three important findings emerge when we examine consumption rather than income. First, there are gaps between what the income statistics say about the resources households have versus what the consumption statistics indicate. For example, in 2004, some 183,000 households reported income of less than $5,000 but these same households’ consumption was, on average, $20,000. The difference is due to some combination of unreported income, monetary or inkind gifts, the drawing down of savings, and foreign sources of income, none of which are included in income statistics.
Second, consumption data consistently show lower levels of inequality compared to income. That is, when we examine what households consume versus their income, we find lower levels of inequality. Third, consumption data show lower increases in inequality over time.
The solutions often advocated from simple analyses more often than not would make matters worse. For example, a consistent recommendation and one included in the recently released Organization for Economic Cooperation and Development report is to increase marginal tax rates on higher earners. In Canada, this means increasing taxes on the top 20 per cent who already pay roughly 55 per cent of the total tax burden while earning 47 per cent of income.
Critically, most of these studies assume that increasing such taxes has no effect on the economy. Indeed, some studies argue that increasing these taxes improves the economy. Yet the data and experience demonstrate the reverse. Jobs are created when existing business expand or new businesses are created. The capital for both must come from savings. Reducing the rewards to savings as well as to the success of expanding or creating a business necessarily reduces the incentives to do either. Robert Carroll and his colleagues, for instance, investigated this very issue and found that increases in marginal tax rates resulted in large reductions in the proportion of entrepreneurs investing in new capital.
The worst possible solution to improve the lot of the bottom 20 per cent is to pursue policies that would reduce job creation. The best chance to improve the opportunities for lowerincome earners is a robust labour market in which they can obtain onthe-job training and experience. For example, in the decade-plus period preceding the 2008 recession, Canadian poverty rates were reduced by nearly 40 per cent owing largely to a booming labour market.
There are a number of other important factors that contribute to inequality such as technological change, the quality of education, and transformations in the composition of households over time. There are also additional considerations needed to fully understand the dynamics of inequality, such as how our income naturally changes over the course of our life as we move from dependent children, to students, to entry-level workers, to experienced workers, to retirement.
Over-simplifying complex problems so they fit neatly on a T-shirt is never a substitute for thoughtful, informed debate on critical issues facing our country.
Jason Clemens is the director of research at the Macdonald-Laurier Institute and a co-author of the award-winning bestseller The Canadian Century.