Canadians have just entered a decade during which their real disposable income will tend to grow at a much slower pace than in the past. Since 2007, Canadian households have pretty well maintained the same average pace of growth in their real disposable income (per capita) as in the preceding quarter of a century, in spite of a recession and a sluggish recovery. This is set to change in the next decade as adverse demographics will depress real income growth per capita. There is little likelihood that faster productivity growth, increased labour force participation or gains in the terms of trade provide more than a very partial offset to adverse demographics. This means that on average the purchasing power (and saving power) of Canadian households is set to progress much more slowly than in the past. In such a context of slow growth in average household income, issues with respect to income distribution among households are likely to become more acute.
Growth in real disposable income refers to the after-inflation growth of household income after personal income taxes are paid and transfers to and from other agents are factored in. There are two approaches to tracing the sources of growth in real disposable income. The first consists in looking at the evolution and relative importance of real wages, salaries and other components of the real income of households. The second consists in looking at the drivers of the growth in real national income and at the factors determining changes in the household share of that growth. Both approaches rest on national accounts identities. In this report we use the second approach because it establishes a direct link to the fundamental underpinnings of income growth and makes it easier to project what might happen in the future.
Thus, in this second approach, growth in real household income may be seen as depending on two sets of factors: the growth factors, i.e. those that drive growth in real national income and which we mostly associate with the supply side of the economy, such as growth in hours worked and labour productivity; and the distribution factors, i.e. those that determine how much growth in real disposable income households derive from growth in real national income, such as the tax and transfer system. As shown in Table 1 below, growth in real disposable income over the last 30 years has been much more stable than growth in real national income (income of Canadian entities) or real GDP (income produced in Canada)1 because of the stabilizing influence of the distribution factors.
Let us have a more detailed look at the nature of the growth factors and the distribution factors before analyzing the data
- Growth in total hours worked: driven by demographic and economic factors as they affect growth in working-age population, the employment rate and average hours worked per worker.
- Growth in labour productivity: driven by growth in physical, intellectual and human capital per worker and by the pace of technological and organizational innovation.
- Changes in the terms of trade (price of exports relative to the price of imports): driven by changes in real commodity prices in world markets. A rise in the terms of trade contributes to a rise in real national income because it increases the volume of real domestic spending that Canadian entities can afford out of their nominal revenues from domestic production. Terms of trade fluctuate over the business cycle so that over short periods they may have strong impacts on real income growth in one direction or another. Over extended periods, terms of trade tend to have a relatively modest effect on real income growth.
- Changes in net investment income from abroad: take account of the revenues that Canadian entities get from and pay to non-residents. They made a negligible contribution to real income growth in Canada.
- Changes in household share of national income: an increase in this share boosts real disposable income relative to real national income. Over the last few decades the household share has shown a significant downtrend likely reflecting both an erosion of the market power of labour as a result of globalization and technological change and a rise of capital intensity in the economy. In Canada, the household share also tends to respond negatively to a rise in real commodity prices.
- Changes in the price of consumption relative to the price of final domestic expenditures (i.e., consumption + investment + government purchases of goods and services): a fall in the relative price of consumption boosts real disposable income (deflated by the price of consumption) relative to real national income (deflated by the price of final domestic expenditures). Over long periods, this factor has contributed very little to growth in real disposable income because the price of consumption has followed essentially the same trajectory as the price of final domestic expenditures. Over short periods, however, it can make a significant difference.
- Tax and transfer system: includes the transfers received from governments by households, such as EI benefits and old age benefits, as well as the transfers paid by households to governments, such as personal income taxes and EI contributions. The tax and transfer system exerts a counter-cyclical influence on household disposable income as it reflects the operation of automatic stabilizers. For example, during an economic slowdown, growth in personal income taxes declines while EI transfers to the household sector tend to increase.
Real Income Growth in the Last 30 Years
As shown in Table 1, 1984-2007 was a period of fairly robust growth in real national income. Total hours worked made a solid contribution as demographics were favorable (the working-age population grew relatively rapidly), the rate of labour force participation generally rose and the trend unemployment rate tended to decline. Labour productivity growth was fairly moderate relative to its pace in earlier post-war decades and somewhat weaker than U.S. productivity growth. Terms of trade improved along with a rise in real commodity prices, especially over 2003-2007. Growth in real disposable income, however, was markedly constrained by an appreciable decline in the household share of national income and a significant drag from the tax and transfer system. Thus, real disposable income grew by only 2.5 percent per year over 1984-2007 while real national income and real GDP progressed by 3.2 percent and 2.9 percent respectively.
Over the 2008-2014 period, which incorporates a recession in Canada, growth in real disposable income barely declined relative to the 1984-2007 period and far exceeded growth in real national income and real GDP. Low growth in real national income arose from a marked slowdown in the growth of total hours worked and labour productivity and from an appreciable fall in the terms of trade. What saved the day for real disposable income growth was that the distribution factors became favorable to households: the household share of national income increased modestly instead of declining as labour income proved far less sensitive than profits to the economic downturn; the relative price of consumption fell, which boosted real disposable income relative to real national income; and the tax and transfer system on net withdrew funds from households at a lesser rate as growth in personal income taxes slowed more than growth in transfers to households. Thus, real disposable income grew by 2.4 percent per year over 2008-2014 while real national income and real GDP progressed by only 1.4 percent and 1.6 percent respectively.
Real Income Projections over the Next Decade
At the outset, one must reckon that the next decade in Canada will likely witness four important changes relative to the previous two or three decades: demographics will slow economic growth in Canada instead of boosting it; world growth and the pace of globalization will be slower; real commodity prices and Canadian terms of trade will be relatively flat on average (at fairly high levels) instead of generally rising; and the Canadian dollar should fluctuate in a broad range centred on US$0.85, and avoid persistent bouts of acute weakness, as in the decade to 2003.
Our base case projections of growth in Canadian real disposable income are consistent with these expected changes. Over the next decade real disposable income per capita is likely to grow not much above half the rate experienced over the last 30 years, i.e. a meager 0.9 percent per year. This reflects the likelihood that growth in real national income will slow considerably and that the household net after-tax share of that income will decline. The projected slowdown in real income growth would largely arise from adverse demographics: slower growth in working-age population because of earlier falls in the birth rate and a decline in the labour force participation rate due to population aging. Along with a projected further decrease in average hours worked per worker, this would substantially slow the contribution of total hours worked to growth in production and real income. One likely mitigating factor is that the labour force participation rates of the persons 55 to 70 years of age are likely to be on upward trends over the next decade as a result of economic pressures and/or incentives to enter or stay longer in the labour force. Our base case projection factors in such upward trends. However, that still leaves a substantial shortfall in the growth of total hours worked relative to the past.
Table 1: Factors Contributing to Growth in Real Gross National Income and Disposable Income (%)
Click here to view the table.
Sources: Calculations based on data from Statistics Canada Cansim matrices 380-0065, 380-0066, 380-0071, 380-0072 and 051-0001. Note that the calculations were made with the data available prior to the release of the Canadian national accounts on May 29, 2015.
One cannot count on future gains in the terms of trade (rising commodity prices) to significantly offset the impact of this slower growth in total hours worked on real national income growth. Terms of trade have boosted growth in real national income over 1984-2007 and depressed it over 2008-2014. Terms of trade, driven by movements in real commodity prices, are difficult to predict because of the uncertain future evolution of the factors that influence both the demand for resources (e.g., growth in China) and the supply of commodities (e.g., technology). Our judgment is that a scenario of flat terms of trade for the next decade balances the risks, especially in view of the fact that real commodity prices are currently at relatively high levels by historical standards.
At the same time as real national income would slow, the household share of that income should decline, albeit at a much slower pace than in the 1984-2007 period partly because the negative impact of globalization on real wages is likely to diminish and labour should become relatively more scarce in Canada due to demographics. As well, the current tax and transfer system on average is likely to have no effect on growth in real disposable income over the next decade, in contrast with a negative effect over 1984-2007. This is because growth in transfers received by households is expected to match growth in income taxes and contributions to social insurance plans paid by households over the next decade, rather than fall short of it as during 1984-2007. Over the 2015-2024 period, growth in household income and payments will likely slow more than growth in transfer receipts relative to 1984-2007. Transfers receipts for old age are expected to increase relatively rapidly because of population aging.
A relatively fast rise in the quality of technology-intensive consumer goods and services over the next decade, if properly captured by the consumer prices compiled by Statistics Canada, could lead to a decline in the relative price of consumption, and hence boost the growth rate of real disposable income relative to the growth rate of real national income. Although this seems a reasonable assumption to make, we have not factored in this possibility in either our base case or our optimistic case because much progress remains to be made in the measurement of the quality of consumer goods and services. Besides, the pace at which quality improvement will proceed over the next decade is highly uncertain. Still, it is worth keeping in mind that the true pace of real income growth over the next decade may well exceed the measured one as a result of relatively fast (but un-measured) quality improvement in consumer goods and services.
That leaves growth of labour productivity as the only factor that could materially support real income growth (as measured) going forward. In our base case we assume productivity growth over the next 10 years to be slightly above its 1984-2007 average (1.4 percent versus 1.2 percent) on the premise that the reduced abundance of labour, especially skilled workers, starts putting upward pressure on real wages and thus incent firms to invest more in labour-saving technologies. This incentive is likely to be reinforced by business expectations that the Canadian dollar should remain in a range around US$0.85 and not fall to the much lower levels seen in most of the 1984-2007 period.
But productivity growth may well not improve over the next decade and indeed may remain at the low average rate experienced since 2007. Moreover, it is quite possible that labour force participation rates for individual age groups will also remain at current levels. This low caseprojection would imply that real income per capita would grow at only 0.4 percent per year compared to the 1984-2014 average of almost 1.4 percent per year (and compared to 0.9 percent in our base case). While we attach a higher probability to the base case outcome, the low case outcome remains a clear possibility.
All in all, we expect adverse demographics to depress growth in real disposable income per capita to low levels over the next decade. This will severely constrain further progress in the living standard of the average Canadian household. One cannot count on a durable run of gains in the terms of trade to significantly cushion the blow. Even a spurt of labour productivity growth large and persistent enough to make a material difference would be a tall order in view of the disappointing experience of the last 30 years. Yet this is what it would take to bring growth in real income per capita over the next decade (optimistic assumptions) on par with the pace experienced over the last 30 years.