Tuesday, March 6, 2012

Investment cycles


As many of my blog readers know, as a macroeconomist I am a maverick as I believe in cycles.  It seems to me (and a few others, although it should be noted we are definitely in a minority), that cyclical behavior is just part of what economists label the “business cycle” and denying its existence or stating that the cycles are caused by shocks simply put you in denial of the facts.  I would go beyond just the business cycle as a stylized fact – I believe there are other irregular cycles that also permeate the macroeconomy, most notably in the shorter term “growth cycles” that we can also observe just by looking at the growth rate.  These shorter term growth cycles originate from a variety of different sources, but most notably is investment ( - a previous blog posting covered the cycles in consumption expenditure), the reason for this being that investment is one of the most volatile components of aggregate spending.

First off, we need to just clarify what we mean by investment expenditures.  Investment as a concept has a very specific meaning in economics which is quite specific compared with the everyday use of the term “investment” in common language.  Investment in economics means spending on plant, machinery, equipment, new commercial property etc ( - what is known as gross fixed capital formation to an economist), but also included in this category in the national income accounts one also finds spending on new residential housing, computers (whether for households or firms) and changes in inventories.  Note that investment in an economic sense does not include purchases of stocks and shares or bonds or spending on education etc.  Although we might use the term investment to talk about these types of expenditures in everyday language, to an economist this is NOT investment spending.

So let’s look at the different cycles that I hypothesize one should find in investment spending and why.  There are 4 cycles that I would hypothesize exist in the data and each would be expected to have a different periodicity and to depend on different macroeconomic variables.  Let’s take each cycle from the cycle that I would expect to have the longest frequency to one that I would expect to have the shortest frequency. 

The longest cycle that I would anticipate one should find in the data is a “technology innovation investment cycle”.  If new technology is introduced then firms are likely to adopt this technology over time, but in more competitive industries this adoption could be quite rapid.  These are cycles that depend on hardware and often require dramatic changes in the workplace.  For example we have gone from labor being the major input in the production process of automobiles, to computers determining speed of the assembly line and ordering of parts but mostly labor involved in actual assembly to computer-controlled robots actually doing assembly.  In each wave of technological change (computers then computer-controlled robotics), there have been longer term costs savings from a lower labor requirement but a substantial upfront investment on plant machinery and equipment.  These cycles should be longer cycles than the business cycle ( - what some economists refer to as “medium-term cycles”), but they are intimately linked with the business cycle, as they can lead to booms in economic output as manufacturing costs fall, and costs elsewhere in the economy should also fall.  The point is that we shouldn’t expect every business cycle to be driven by these technological innovations, but of course the business cycle will be affected when these changes happen.

The next longest investment cycle has to do with the business cycle.  As consumption is cyclical, any downturn in the economy will result in a fall in consumption which will lead companies to pare back expansion plans, cancel those plans or indeed reduce production capacity.  This will result in a fall in investment as equipment wearing out will not be replaced and new equipment orders will be cancelled. This “demand-led investment cycle” should lag the business cycle.

The third investment cycle also has to do with the business cycle as well, but originates in Keynesian and Austrian thinking.  The real interest rate determines how costly it is to invest and determines the volume of investment according to a negative relationship between the cost of borrowing (the real interest rate) and the amount of investment undertaken.  This could happen through 2 different channels: i) the Austrian school channel which states that the interest rate is inversely proportional to the average length of the production process so that lower real interest rates lead to longer production times ( - and therefore a greater amount of capital required to produce the same amount of output at a given time in the future – a relationship known as the Hayekian triangle); and ii) the Keynesian school that simply states that lower real rates make it cheaper to borrow and thus more firms should borrow in order to expand.  Either way this implies that monetary policy will have an impact on investment and indeed exogenous shocks that lead to low or expectations of negative real rates would create incentives to increase investment.

The fourth investment cycle is of variable length but has to be with depreciation cycles.  If historically there were cycles in investment for whatever reason (“technology innovation” or “demand led” cycles) then these cycles will likely replicate themselves as this equipment will eventually wear out and need replacing.  Presumably the equipment would need replacing at roughly similar times given the fact that it was installed at roughly the same time, so this will lead to “upgrade” programs to replace old equipment and this tends to occur over a relatively short period of time relative to the business cycle.  This I will call “depreciation investment cycles”.  Obviously if we assume technology to be fixed, the depreciation investment cycles will depend on the quality of the capital and the availability of parts.  An obvious example of this would be aviation or transportation equipment that usually is used until it wears out and parts no longer exist or are prohibitively expensive.

The fifth investment cycle relates to the relative price of capital, and this is usually measured by something called “Tobin’s q” theory.  If you think of companies as using factors of production labor and capital, then the valuation of a company is usually done based on it’s “assets” which, ignoring  labor force skills is calculated by the value of the physical assets (capital) which would remain if the company ceased to operate.  Comparing the value of the company (as measured by the cost of replacing its assets) versus how much the stockmarket currently values the company (how much the company is worth in terms of future profitability) gives a relative value of capital and is known as “Tobin’s q” theory after the Yale economist James Tobin.  Any calculation of “Tobin’s q” (Stockmarket value to replacement value) appears to move in cycles, and therefore implies that at certain points the stockmarket values capital more highly than the cost of capital and therefore there is an incentive to expand by increasing investment. 

The last investment cycle I will outline is a “competition-driven investment cycle”. In competitive economies with industry structures that resemble anything other than monopolies, there will be pressure to “keep up with the Jones’s” in order to maintain market share.  Although competition has a role to play in the “technology innovation investment cycle”, it also has a role to play in terms of offering a similar consumer experience as competitors do, and this might require software or equipment upgrades to meet consumer expectations in a competitive marketplace.  Examples of this type of investment may well take place by “third party” suppliers such as internet service providers for banks, modern d├ęcor in coffee shops or the look and feel of a supermarket.  This “competition-driven investment” is likely to occur over a discrete period of time between the different firms in the marketplace as firms attempt to differentiate themselves from one another. 

OK, enough theory.  The proof of this approach will clearly be in the pudding. More next time on the relationship between stockmarkets, economic growth and investment. 

Featured Post

Free Trade on Trial - What are the Lessons for Economists?

This election season in the US there has been an extraordinary and disturbing trend at work: vilifying free trade as a "job kille...

Popular Posts

Search This Blog