Canada’s Underperforming Corporations
In neoclassical economic theory, corporations are supposed to “work”, just like the rest of us do.Â Their economic function is to organize production, innovate, and grow.Â This process, when it happens well, generates jobs and incomes (which is not to say there are not better ways to generate jobs and incomes).Â One way to measure the “work” of corporations is by their investment: how rapidly are they mobilizing both financial capital and real resources into new, value-producing, job-creating projects.
By this measure, Canada’s corporations are downright lazy these days.Â Their investment performance has been ho-hum at best in recent years — especially when compared to the unprecedented run-up in corporate profits.Â They are being paid more, to do less.Â And as a result they are quite literally sitting on top of more money than they know what to do with: non-financial corporations are now net lenders to the rest of the economy (when they should be net borrowers, if they were indeed doing their “job”).
I was recently asked to provide some background data on investment spending by Canadian businesses.Â Here is the memo I prepared.Â The background data is available from me by e-mail (email@example.com), and any feedback is most welcome.Â My report also contains several graphs that don’t convert well to this blog page; those are also available by request via my e-mail.
As you will see, I think this issue provides an excellent opportunity for the left to challenge the logic and the legitimacy of the whole business-friendly approach to economic policy that has dominated Canada and other countries in recent years:
Business investment spending in Canada has been very weak in recent years â€“ in absolute terms, but especially relative to the unprecedented improvement in business profits.Â As a share of GDP, total business non-residential investment spending has increased only slightly since the low point of 2003. Â All of this increase in share (and half the increase in spending in dollars) was due to booming investments in the oil & gas and mining sectors.
Business spending on machinery and equipment (which many economists think is the most important form of investment) has fallen as a share of GDP since 2002 (when the rise in the Canadian dollar started to make imported machinery less expensive for Canadian purchasers).Â This disproves the claim that a higher dollar will enhance Canadaâ€™s productivity by encouraging companies to invest in new technology; the cost of that technology (in Canadian dollars) may have fallen, but the rationale for locating it in Canada has eroded badly because of the strong dollar.
Since 1999, barely half of the new profits received by Canadian businesses have been reinvested in fixed capital assets in this country.Â The combination of historic profits with weak reinvestment has resulted in a large financial outflow from real Canadian businesses â€“ which in economic theory are supposed to be net users of financial capital (provided by banks, the stock market, and other financial intermediaries).Â In 2006, non-financial business accumulated about $40 billion in surplus cash.
By no stretch of the imagination can it be argued that business investment in Canada is currently â€œstrongâ€ â€“ whether relative to our history, relative to the needs of our economy, or relative to the all-time profits being captured by businesses.Â The failure of business to reinvest its record profits is a powerful refutation of the logic of globalization, corporate tax cuts, and the other pro-business measures which have been implemented in Canada.
The discussion below relies on a spreadsheet I have constructed that includes quarterly data on business fixed investment and related variables going back to 1961, from CANSIM.Â In addition to the usual GDP components, the spreadsheet also includes financial flow data from the sector accounts which CANSIM also provides.Â This data provides additional detail on the non-financial business sector (which is of particular interest for us), whereas the GDP data describes only an aggregate business sector.
For this discussion, we are interested in non-residential business investment (we all know there has been a housing construction boom, that is likely in the process of ending).Â Of course, public investment is a crucial priority, too, but it is determined by separate factors; our discussion here focuses on business investment.Â We are interested in fixed investment, not inventories (which exhibit normal, ongoing cyclical fluctuations).Â In some cases, we are particularly interested in investment in machinery & equipment (excluding structures), which is thought by some to be especially important to productivity growth and to be a leading indicator of overall macroeconomic conditions (since M&E commitments are considered a particularly sensitive indicator of business confidence and intentions).
The spreadsheet also provides some annual breakdown on energy and non-energy fixed investment.Â Investment data by sector is not available on a quarterly basis, only an annual basis (through StatsCanâ€™s Private and Public Investment publication and related CANSIM series).Â The 2006 data from this series is â€œintentionsâ€ â€“ that is, what companies thought they were going to invest in 2006, when StatsCan asked them early in the year.Â This data is important for determining the extent to which recent growth in investment spending reflects the oil sands boom in Alberta.
One issue that comes up is the role of deflators in understanding â€œrealâ€ investment effort.Â There is a real investment series in the constant-dollar GDP-by-expenditure table that often grows faster than nominal investment spending.Â This strange result is due to the impact of rapid technological change in computers and computer-related equipment (which is an important part of business M&E investment), and also to the impact of currency fluctuations on the prices of imported technology.Â It is notoriously difficult to reliably impute the genuine impact of falling computer prices (while also adjusting for quality, etc.); this is a long-standing problem in calculating price deflators, estimating productivity growth, and other issues.Â I do not use this real investment series.Â While it may show, in some cases, that the growth of â€œreal investment effortâ€ seems to be better than the growth of actual investment spending, it is still an imaginary and unreliable statistic.Â Moreover, it sheds no light on what businesses are actually doing with their money (or, more broadly, what society is doing with its resources).
Even worse, sometimes investment effort is measured by the share of (constant-dollar) investment spending in (constant-dollar) GDP.Â This provides an even stranger result, since one (suspiciously) deflated series is divided by a denominator that is deflated by a completely different price index.
It is more reliable to measure the strength of investment by the growth of actual investment spending, in dollars.Â In most cases, we also want to compare that to the growth of the overall economy (GDP) or business financial indicators (profit, cash flow, etc.).
In general we measure gross investment spending, which is the amount of new investment spending actually injected by businesses.Â This includes the amount of spending required just to offset the wear and tear of existing capital assets (depreciation).Â Again, it is the injection of actual dollars that drives the economy, so the gross value is more relevant for macroeconomic performance, job-creation, etc.Â Also, it is rare that businesses replace used-up capital equipment with exactly the same machinery; usually they update the technology involved.Â So for understanding technological change and productivity growth, the gross value is best.Â Sometimes, however, the net value (after deducting expenses for current depreciation) is more relevant â€“ for example, if we want to estimate the rate of growth of the overall (net) capital stock, or if we want to compare investment to some financial flow (such as bottom-line profit) from which depreciation costs have also been deducted.
Anyone wanting more detail on how to measure investment could read Chapter 7 of my 1999 book Paper Boom.
Recent Investment ResultsÂ
Business fixed investment spending has been increasing at a moderate pace in recent years.Â For all sectors, business non-residential fixed capital spending has been growing by about 8-9% per year since 2004.Â This is somewhat faster than the growth in (nominal) GDP, so the share of this variable in total GDP has grown somewhat (from a low of 11 percent in 2003, to about 12 percent at present).
All of this growth in the investment share has been due to growth in investment in structures.Â Investment in business machinery and equipment (M&E) has been much less vibrant â€“ growing at about 5% per year beginning in 2004.Â This growth slowed notably in 2006 (to about 3%).
What about the trend in both investment variables measured as a share of GDP?Â As mentioned, non-residential fixed investment spending has climbed to 12% of GDP.Â This offsets half of the downturn in business spending which occurred between 1999 (a cyclical peak for
Canadaâ€™s economy) and 2003.
On the other hand, machinery and equipment spending has not increased at all as a share of GDP â€“ hovering at about 6.5%.Â The appreciation of the dollar has had no visible positive impact on the share of GDP allocated by business to new technology.Â (Note that investment in structures should not be sensitive either way to the value of the dollar, since it consists mostly of construction spending which is mostly a non-traded activity.)
The fact that all of the growth in business non-residential fixed investment (as a share of GDP) has been due to investment in structures, is already an indication that the massive energy and mineral developments which have recently been sparked by high global commodity prices are central to the recent expansion of overall business investment.Â Outside of the resource sector, most businesses are loathe to invest heavily in new structures â€“ which are often disparaged (by managers and financial analysts, alike) as â€œbricks and mortar.â€Â Spending on technology is considered more vital to longer term economic success than new buildings, yet this type of business investment has been stagnant.
The contrast between the general stagnation of investment and the historic growth of business profits is striking.Â Corporate profits (measured after income taxes and net of government subsidies) are at all-time record levels in Canada.Â After-tax profits have more than doubled since 1999, and have grown by half as a share of GDP (from 8% to 12%).Â Before-tax profits are also at an all-time record; but the positive impact on the bottom line has been reinforced by big corporate tax cuts enacted during this time.
It is interesting to note that in 1999 after-tax corporate profits were roughly equal to the amount of business investment in machinery and equipment.Â Today they are roughly equal to the total sum of non-residential investment.Â Remember: profits are net of depreciation, whereas gross investment spending is not â€“ so in general, the flow of profits should be much smaller than the flow of gross investment.Â What we see today is a strange process of net lending from non-financial businesses (which are earning record profits, but reinvesting a diminishing portion of them) back to other sectors of the economy.Â In 2006, this net lending equalled about $40 billion.Â By the hard numbers, businesses clearly have more money than they know what to do with.
The weak investment performance of business is further highlighted by measuring investment spending relative to business finance.Â There are two ways to do this: one gross, and one net.Â The gross ratio compared gross fixed investment spending to gross business cash flow (before the deduction of non-cash charges against profit, mostly depreciation).Â The net ratio compares net fixed investment (after depreciation) to bottom-line profit (also net of depreciation).Â For the non-financial business sector, companies are re-investing only 80 percent of their available cash flow (and 60 percent of their net profit) back into new Canadian projects.Â Traditionally, in a growing economy, both of these ratios should be well over 100 percent (with the extra needed cash provided by the financing activities of banks, stock markets, and other financial institutions.)
That the phenomenal growth in business profitability has not translated into new business investment spending can be dramatized as follows:Â Compare the dollar growth in business profits since 1999, to the dollar growth in business investment spending.Â Two measures are again considered: one from the GDP accounts (for the whole business sector), and one from the financial flow accounts (for the non-financial business sector only).Â In both cases, it appears that every dollar of new business profit since 1999 has resulted in about 55 cents of new business investment spending.Â Put differently, we have had to reward the business community with almost $2 in new profit since 1999, for each $1 in new investment spending that they have grudgingly pumped into the Canadian economy.Â Thatâ€™s not much bang for the buck.
Everyone agrees that the energy and minerals investment boom has been a major source of new investment spending in the Canadian economy.Â The lack of quarterly data on investment by sector makes it tricky, but not impossible, to show this.Â We can map the annual sector investment data from StatsCanâ€™s investment intentions survey onto the quarterly series from the GDP accounts, in order to subtract energy and mining investments from the total.
One caveat to this approach: the 2006 energy & mineral spending numbers are intentions only; no actual data is available yet for 2006.Â On the other hand, the total investment numbers are actuals (for the first three quarters of the year).Â If energy investment spending turns out to be greater than initially intended, then estimated non-energy spending will be even lower (and vice versa if energy spending turns out to be weaker than initially expected).
By this approach it is clear that all of the growth in total non-residential fixed business investment (measured as a share of GDP) since the low point of 2003 has been due to energy and minerals projects.Â In fact, non-energy fixed investment is lower in 2006 than in 2003 as a share of GDP (and grew only very slightly in 2006).Â Non-energy fixed spending will equal about 8.5% of GDP in 2006 â€“ compared to 9% in 2002 and over 10% in 1999.
In dollar terms, the energy sector accounts for almost half of all the new fixed non-residential investment spending by business since 2003 (even though energy and minerals traditionally accounted for only about a fifth of all business capital spending).Â Without the energy and minerals boom, it is clear that business investment spending would look even weaker than the aggregate numbers indicate.Â Many Canadians, of course, are concerned about the long-run economic and geopolitical implications of
Canadaâ€™s re-specialization as a producer and exporter of unprocessed resource-based commodities.Â These investment numbers indicate powerfully that this re-specialization is indeed occurring; before our eyes, the flow of profit-seeking private capital is recreating Canada as a â€œhewer of wood, drawer of water.â€
Implications and AnalysisÂ
In my view, the apparent weakness of business spending is an unanswered question.Â Why has business investment been so weak, especially measured relative to improved profitability?Â I have traditionally believed (like many left economists) that profit rates are indeed a significant determinant of new investment; my 1999 research for Paper Boom suggested that this relationship did indeed hold in
Canada (along with other determinants, like capacity utilization and interest rates).Â Today, clearly, the link between profits and investment is much weaker.Â This is very damaging to the whole logic of trickle-down economics: if higher profits do not result in stronger investment spending, then reducing costs (whether wages, business taxes, or other costs) merely redivides the pie in a less equal fashion, and does little if anything to increase the size of the pie.
I suspect the post-2003 weakness in business spending has much to do with:
- the structural impacts of globalization (especially the increased import penetration from low-cost emerging economies)
- the impact of the dollarâ€™s appreciation (I imagine the high loonie had a negative effect that helped to wipe out any positive effect from higher profits â€“ producing the apparent stagnation in the investment share since 2003)
- the large proportion of business profits that has been concentrated in the resource sector
- the ongoing financialization of the economy (another good portion of business profits are located in
Canadaâ€™s overdeveloped finance industry)
It will take further work to clarify these and other potential explanations for the weak Canadian business investment performance.Â It is worth keeping in mind that business investment has also been weak in other OECD countries, especially again relative to business profits and cash flow (a fact that has been noted in recent studies by the U.S. Federal Reserve, the IMF, and others).Â This would lend more support to the theory of weak investment reflecting a structural shift resulting from globalization.
While we are working to better understand this weak investment performance, we should definitely make the failure of business to reinvest its profits in Canada a major theme in our general critique of neoliberalism.Â The extremely weak link between profits and investment breaks the logical chain of trickle-down policy (a chain which was â€œiffyâ€ to begin with).Â And it is a big crack in the credibility with which the business sector poses as the efficient managers of our economic development.Â They are receiving each year tens of billions of dollars in surplus profits which they literally do nothing with, despite the obvious needs of Canadians for more investment (in both the private and the public spheres).
If they refuse to put that â€œmoney into motionâ€, in productive and welfare-enhancing ways, then some other economic player or players should be given the job instead.Â We should take back those excess profits (by rescinding, for example, the no-strings-attached business tax cuts which have cost so much since 2001), and use the money instead to fund public and non-profit investment projects.Â And at the same time, we should implement policies which stimulate business investment (which we clearly still need) in more powerful, direct ways â€“ like â€œpay-or-playâ€ ITCs, pro-active industrial policies, performance-related investment subsidies, and strategic trade policies.
We’ve gone through this before. You’re using the series for nomininal expenditures on investment and nominal GDP. When you look at the data in Table 380-0002, you’ll see that real investment as a share of real GDP is at an all-time high, and that this is being driven by machinery and equipment.
We know that the relative price of M&E has fallen: most of it comes from the US, and the appreciation of the CAD has made M&E relatively cheaper. The fact that nominal expenditures have gone down as a fraction of nominal GDP just means that the demand elasticity for M&E with respect to its price (leaving aside all the other things in the cost of capital) is less than one.
Real investment – the thing that actually matters in a discussion of capital accumulation and productive capacity – is at an all-time high, in both absolute terms and as a share of real economic activity.
Stephen, it seems to me that Jim convincingly pre-empted this argument in the third through fifth paragraphs of “preliminaries.” To measure the proportion of Canada’s economic resources devoted to investment, it makes sense to divide investment by GDP. I am not sure what dividing investment by one price index, dividing the result by GDP, and then multiplying the result by a different price index tells us. Your point, I suppose, is that the unimpressive levels of investment spending identified by Jim may make impressive contributions to the capital stock, given that some capital is getting cheaper. The fact remains that Canada’s economy would be much stronger if investment spending rose to more traditional levels.
Nominal expenditures = price * quantity.
For the purposes of a discussion of investment and productivity, quantities are what matter. Not nominal expenditures.
I agree that, to measure growth of the capital stock, we want to look at quantities as opposed to expenditure. Deflating expenditure with a price index, as you favour, may approximate “quantities”.
However, to measure investment flows relative to Canada’s economy or to business finance, expenditure is the relevant variable.
Not at all.
Consider the case of a household that buys only apples and oranges, and that it spends an equal amount on both. Now suppose that the relative price of apples falls. The household will buy more apples and fewer ornages, but there’s no reason a priori to think that total expenditures on apples will rise. We’d see more apples and fewer oranges, but the expenditure shares can go either way, depending on the demand and/or substitution elasticities.
But Jim’s argument was not that, because the relative price of capital goods (apples) fell, expenditure on capital goods (apples) should have increased.
Only if the price elasticity is greater than one. Quantities went up, but since the rate of change of quantities is less than the rate of change in prices (i.e., the elasticity is less than one), total expenditures went down.
There’s no reason to expect that the demand elasticity for M&E should be greater than one.
Call me simple-minded, but corporate profits are earned in nominal dollars. If the investment effort is lagging in terms of nominal dollars, then the reward/rate of return to capital in nominal dollars is higher than needed – as Jim concludes we should be able to tax more and re-direct surplus profits to better uses.
Stephen, I think that you may have missed the word “not” in my last post.
Quite right; sorry, Erin.
Pingback: My Blahg » BLAHG BITS: JAN.18/07
A few rejoinders to the earlier discussion:
1. I am skeptical of the meaningfulness of any of the price deflators which attempt to capture changes in the “real price” of computer equipment. These used to produce utterly bizarre findings (eg. they calculated quality-adjusted price indices based on the “real” price of 64 KB RAM worth of computing power — and found that the real price of computers had fallen near to zero). Modern techniques (eg. chain-linking) are better, but still produce wierd results when tech change is fast. Exclude computers from the M&E price index and it is not at all clear that capital prices are falling.
2. And generally, any decline in a quality-adjusted “real” price is irrelevant for business purposes — because any business which wants to compete successfully needs to have the cutting-edge technology anyway. (Another way of putting this: because of technological change and competition, companies have to invest MORE, in so-called “real” terms, just to maintain their competitive position.) The fact that the “real” price of a 64-KB RAM computer is next to zero, won’t save the skins of any business executive who chooses to “buy” them. In reality, they have to spend as much or more as they ever did on computers, despite their falling “real” price, to keep up with technology and the competition.
3. There are also issues in how StatsCan calculates the import price index for specialized capital goods, for which there is no transparent and easily-measurable “market” to look at. In many cases they simply deflate the former price by the change in the exchange rate. But we know from consumer price indices (which are more transparent and hence reliable) that this is utterly invalid: foreign producers, importing middlemen, & retailers capture much or most of the cost savings resulting from a strong dollar in the form of much fatter margins (just as they were forced to “eat” most of the cost of the dollar’s former fall in the form of thinner margins). We know, contrary to conventional wisdom, that currency fluctuations have almost no impact at all on final consumer prices in Canada, and I suspect the same is largely true of capital goods.
4. For both reasons, I am very skeptical of the claim that investment goods are cheaper, in any meaningful sense. This is consistent with my anecdotal experience from the auto industry and other industries in which I represent the CAW.
5. Further to Andrew’s comment: corporate profits (and non-cash charges, like depreciation) are measured in nominal dollars (I’d like to see an investment analyst satisfy his clients with an argument that corporate profits are actually larger than they “seem,” in relation to some hypothetical price deflator!!!). Nominal profits are way up, but it costs less to undertake a given “real” investment effort (if you buy this argument). In this case, there’s a huge rent being skimmed off by corporations and their owners. Here’s where the apples vs. oranges anaology breaks down: an individual consumer can choose for themselves how to allocate the resulting “price” and “income” effects, but most Canadians have no say at all in deciding whether these “windfall savings” resulting from “cheaper” capital prices are used to pay for additional capital investment — or are instead socked away in dividends and corporate cash accumulation.
6. If “real” spending on M&E is at all-time record levels, then why is productivity growth (which was clearly empirically dependent on M&E spending in the past) near zero?