Statistics Canada
www.statcan.gc.ca
Common menu bar links
Capitalization of Software in the National Accounts
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.
4.0 Software investment in Canada and the U.S. compared
Many observers will no doubt be interested to compare the software results for Canada with those of the U.S. Charts 14-23 give some comparisons for 1981-2000, based on the Canadian software estimates as of 31 May 2001 and the U.S. estimates as of 27 July 2001.12 All but the last two Charts show ratios of some variable for Canada relative to the same variable for the U.S.; when a ratio equals one, there is equality between the two countries in the variable compared.
Software investment in Canada, in nominal terms, outpaced that in the U.S. over 1981-88, 1993-94 and 1999-2000, while it led in the U.S. over 1989-92 and 1995-98. Over the entire period, growth in total software investment was slightly higher in Canada. By 2000, software investment in Canada was 14.0 times its level in 1981, while in the U.S. it was 13.2 times its 1981 level. Looking at the different types of software, own-account grew at roughly the same pace in Canada as in the U.S. until the early nineties but has grown faster in the U.S. since then. Spending on pre-packaged software grew slightly faster in the U.S. over the entire period, while the growth of investment in custom software has been much greater in Canada, especially during the eighties.
In 1981 the Canadian and U.S. investment mixes by type of software were markedly different. The share of customized software in total software investment expenditure in Canada was only one-half (0.5 times) its share in the U.S. The shares of pre-packaged and own-account software, on the other hand, were 1.7 and 1.2 times their respective shares in the U.S. These differences narrowed throughout the eighties, and while they have widened somewhat since then, the investment mixes were still more alike in 2000 than in 1981. By 2000, the shares of custom, pre-packaged and own-account software in Canada were 0.75, 1.5 and 0.9 times their respective shares in the U.S. This convergence was driven mostly by the trend of software investment expenditure in the business sector. Canada/U.S. software spending patterns in the government sector in contrast are quite different and have shown little tendency to converge.
Businesses account for roughly the same, growing share of total software investment in both countries, three-quarters of the total in nominal terms in 1981 and four-fifths in 2000. Business sector shares by type of software, however, are notably different, especially for custom and own-account software. This stems in part from differences in the software spending of governments. The relatively low share of total custom software investment accounted for by U.S. businesses compared to their Canadian counterparts (67% versus 94% in 2000), reflects the relatively large market impact of U.S. government spending on highly specialized software for space and defence applications. In the case of software developed on own-account, the relatively low share of the total accounted for by Canadian businesses compared to their U.S. counterparts (67% versus 84% in 2000), partly reflects the greater tendency of governments in the U.S. to contract out to the private sector.
Software prices in both countries evolved in tandem over 1981-2000, with a steep decline for pre-packaged software, in quality-adjusted terms, a relatively flat profile for custom software and a near doubling of own-account costs. The relative (Canada-U.S.) price for pre-packaged software was only 5% higher in 2000 than in 1981, following an 81.7% decline in Canada over the entire period, versus a slightly greater 82.5% decline in the U.S. Prices for pre-packaged software in Canada fell more slowly than in the U.S. during the early eighties, more rapidly during the late eighties, and more slowly again through most of the nineties, reflecting movements in the exchange rate.13 The relative (Canada-U.S.) price for own-account software stood about 6% lower in 2000 than in 1981, reflecting slower growth in the earnings of programmers and systems analysts in Canada compared to their U.S. counterparts, especially from the mid-eighties to the mid-nineties. The relative price for custom software moved in line with that for own-account.14
As a share of GDP and Gross Fixed Capital Formation (GFCF), in nominal terms, software investment in Canada has fallen well below its share in the U.S. throughout 1981-2000, an indication perhaps that Canada has lagged in terms of the computerization of the workplace.15 In 1981 software investment accounted for only 0.3% of GDP in Canada, one-half its 0.6% share of GDP in the U.S. Since then software investment has grown against GDP in both countries, and Canada has narrowed the gap (at least in relative terms). By 2000, software investment accounted for 1.4% of GDP in Canada, nearly two-thirds of its 2.4% share in the U.S.
The growth of software investment in real, inflation-adjusted, terms has been nothing short of phenomenal. Over 1981-2000, software investment grew an average 4.2% per quarter in Canada compared to 3.7% in the U.S. Canada’s edge during this period, is mostly attributable to its lead over the eighties, one that subsequently disappeared. During the 1980s, software investment in Canada outpaced that in the U.S. on average by 1.2 percentage points per quarter, with 5.1% quarterly growth in Canada versus 3.9% in the U.S.16 The recession of the early nineties, which impacted the Canadian economy more severely, brought the growth in software investment down to 3.1% per quarter in Canada and 3.0% in the U.S. With the high-tech boom and software upgrades, replacements and preparations for Y2K in the latter half of the nineties (1996-2000), software investment regained momentum, rebounding to 4.2% per quarter in the U.S., slightly above the 4.0% growth posted in Canada.
Software investment contributed an average 0.03 percentage points to quarterly GDP growth in Canada and the U.S. over 1981-2000. With its growing share of GDP and the quickened pace of investment over 1996-2000, its contribution to GDP growth stepped up in both countries. Over this period, software investment contributed an average 0.05 percentage points per quarter to GDP growth in Canada, somewhat below its 0.06 percentage point contribution in the U.S.
| Table of contents |
