The Paradox of Investment: A Contribution to the Theory of Demand-Led Economic Growth

ABSTRACT

This paper has two purposes. The first is to argue that aggregate investment may be subject to the following paradox. A rise in investment decided by firms to correct overutilization of their production capacity may generate less capacity than demand — and hence cause a paradoxical rise in overutilization. This will in turn lead to even more investment, and so on — the result being the self-sustained rises in output that characterize economic expansions. The second purpose of the paper is to put forward one reason why the above paradox of investment will lose strength as expansions progress, and may eventually disappear leading to their end. That reason may be summarized as follows. As net investment increases along expansions, the effect of investment on production capacity rises relative to its effect on demand — and, as a result, the rise in utilization slows down. Moreover, as net investment eventually grows to a high level, the effect of investment on capacity may become bigger than its effect on demand. If this happens, utilization will stop rising and start falling, and thus the same may happen with investment and output.

KEYWORDS:

• Economic dynamics
• paradox of investment
• economic growth
• business cycle

JEL CLASSIFICATION:

• E12
• E32
• O41

Acknowledgments

We are grateful to two anonymous referees for comments. We also thank Amitava Dutt, Daniele Girardi and Riccardo Pariboni for an exchange of ideas about some of the issues discussed in this paper. Needless to say, all errors are our entire responsibility.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Notes

1 In an economy without government (hence τ = 0) and with no saving out of wages (c_w= 1), the multiplier would be reduced to the more familiar 1/(s_p.π), s_p denoting the marginal propensity to save out of profits.

2 From January 1967 to May 2021 the average of the total index of capacity utilization in the U.S. was 80.03 percent (see https://fred.stlouisfed.org/series/TCU).

3 This mechanism can be extended to housing investment. In this case, the story may be told as follows. As the number of unsold houses — the ‘amount of spare capacity’ — falls below the desired level, building firms increase construction above the replacement level in an attempt to raise the number of houses available for sale to the desired rate. But this increases incomes (firstly in the construction sector) and therefore aggregate demand. If the value of the new houses built — the additional ‘capacity’ — is lower than the rise in aggregate demand, average utilization across the economy will rise.

4 Note however that some heterodox economists have argued that in the long-run aggregate demand growth is essentially driven by autonomous trend rises in private consumption, housing investment, government expenditure and exports (Girardi and Pariboni 2016). We will return to this issue.

5 These last two values are roughly in line with the empirical evidence presented in the preceding section. Note also that, to keep the example simple, we neglect the fact that capital accumulation will imply increasing amounts of capital depreciation, and assume this fixed at $100.

6 Our model can thus be seen as a complement to the super-multiplier model: this focus on the long-run trend of the economy while we try to account for the medium-term fluctuations around that trend.

7 For the period 1960–2011 in the U.S., Trezzini (2011, p. 583) calculated an elasticity of consumption with respect to net disposable income equal to 1.73 on average during economic expansions, and always lower than 0.4 during recessions.

8 For instance, and as shown in the row of Table 1 related to t = 1, the $10 increase in gross investment earlier in the expansion from $100 – the replacement level – to $110 led to a net investment of $10 and to an increase in capacity of $10*(1/10). But, as shown in the row related to t = 10 of that table, the same $10 increase in gross investment later in the boom, from $190 to $200, translates into a bigger net investment, $100, and into a bigger increase in capacity, $100*(1/10).

9 Given the assumed replacement level of investment of $100, a gross investment of $260 corresponds to a net investment of $160 and thus raises capacity by $160*(1/10).

10 For example, the Obama fiscal stimulus package of 2009–10 may have been the trigger that started the recovery from the Great Recession of 2008–9.

Additional information

Funding

This work was supported by FCT – Fundação para a Ciência e aTecnologia (Portugal) under the research [grant number UIDB/05069/2020].