Putting Austerity to Bed: Technical Progress, Aggregate Demand and the Supermultiplier

ABSTRACT

The paper investigates the determinants of private investment and economic growth from a theoretical perspective. We start with a critical analysis of the crowding-out effect and we present a new version of the Sraffian Supermultiplier: a model that accounts for both the multiplier and accelerator effects. We focus on different types of fiscal policies: generic ones and ‘mission-oriented’ ones that set a new direction for the economy. We show that mission-oriented policies have the potential to generate the largest positive effect on investments and output growth as well as on innovation processes and labour productivity growth.

KEYWORDS:

• Mission-oriented innovation policy
• Sraffian supermultiplier
• austerity
• labour productivity

JEL CODES:

• E22
• E60
• O25
• O30

Acknowledgements

We warmly thank all the participants of the EAEPE Conference, held at Corvinus University of Budapest in October 2017, for feedback received on a preliminary draft of this article. The authors would also like to thank two anonymous referees and the following people for fruitful discussion: Riccardo Pariboni, Santiago Gahn, Enrico Sergio Levrero, Luigi Salvati, Antonella Stirati, Francesca Tosi and Attilio Trezzini. All errors remain our own.

Disclosure Statement

No potential conflict of interest was reported by the authors.

ORCID

Matteo Deleidi http://orcid.org/0000-0001-6371-5255

Notes

1 In the case of a negative natural interest rate under a zero lower bound condition or a liquidity trap situation, NK authors recognize that monetary policy is ineffective to stimulate aggregate demand.

2 A recent stream of literature in macroeconomics, by rehabilitating the concept of ‘Secular Stagnation’ (Summers 2014), affirms that protracted and prolonged falls in output can influence also the potential output (Fatás and Summers 2018). In particular, cyclical conditions – for instance, a fall of aggregate demand – lead to permanent effects on both aggregate supply and potential output by means of a mechanism that is typically termed ‘hysteresis’ (Yellen 2016).

3 The NK perspective seems endorsed by a prominent institute like the IMF, which has recently criticized the EA narrative. Episodes of fiscal consolidation have led to a fall in output, an increase in the unemployment rate and an increase of the debt-to-GDP ratio (Guajardo, Leigh, and Pecatori 2011; Ostry, Loungani, and Furceri 2016). According to IMF, expansionary fiscal policies are effective in fostering GDP, investment and employment growth (Leigh et al. 2010; Guajardo, Leigh, and Pecatori 2011). In a recent empirical study, Blanchard and Leigh (2013) show that the fiscal multiplier assumes a positive value equal to 1.5 suggesting that a fiscal consolidation generates a Keynesian effect, thus causing an economic recession rather than an expansion. Yet, despite IMF economists support the positive Keynesian effects of an expansionary fiscal stimulus in the short-run, they maintain that fiscal consolidation is beneficial in the long-run, as this lowers public debt and subsequently reduces the interest rate and distortionary taxes (Leigh et al. 2010).

4 In this paper, we criticise the expansive austerity measures by discussing the direct relationship between interest rate and investment. However, additional mechanisms supporting the expansive austerity measures should also consider the role of consumption in light of the well-known ‘Ricardian Equivalence’. For an in-depth theoretical and empirical review of the inconsistency and weakness of the EA measures, see among others Auerbach and Gorodnichenko (2012), Boyer (2012), Jordà and Taylor (2016), Botta (2018) and Fatás and Summers (2018). Additionally, for a discussion of the ‘Ricardian Equivalence’ and its underlying assumptions, see among others Romer (2007), Ciccone (2013) and Hayo and Neumeier (2017). Additionally, for a monetary critique to these views, namely a discussion on the idea of an endogenous rate of interest, see among others Moore (1988), Rochon (2001) and Lavoie (2014).

5 A recent lively debate on the role played by autonomous components has been developed by Allain (2015), Lavoie (2016), Hein (2018) and Palley (2019), among others. The notion of the supermultiplier and the role played by the autonomous components of aggregate demand in affecting the economic growth was also shared and acknowledged by contributions carried out by Kaldor (1940, 1970, 1975, 1989). Kaldor and his followers focused on the role of export and the external demand (Kaldor 1970) and Thirlwall (1979).

6 For an in-depth review of the concept of autonomous expenditures, see the recent special issue published in Metroeconomica entitled ‘Autonomous Demand, Capital Utilization and Economic Growth’. Particularly, see among others, Dutt (2019), Fiebiger and Lavoie (2019), Palley (2019) and Serrano, Freitas, and Bhering (2019).

7 When $X$ is equal to 0, the banking system does not credit-constrain borrowers. Conversely, the closer $X$ is to one, the more the banking system tends to constrain borrowers. The trustworthiness of borrowers can be measured in terms of, for example, banks’ collateral request.

8 Both total taxes and the total transfers depend both on an autonomous component and on an endogenous component related to wage and profit shares.

9 Following Girardi and Pariboni (2016), we assume that actual degree of capacity utilisation is equal to the ratio between the actual level of output and the normal level of output. Hence, the normal degree of capacity utilisation is equal to one. Subsequently$v=K/y^n$, where $K$ is the actual capital stock and $y^n$ is the normal level of output desired by entrepreneurs.

10 We define the degree of capacity utilisation as the ratio between actual and normal output. It follows that normal utilisation is equal to 1.

11 Since some fluctuations of demand could not be considered permanent, entrepreneurs do not immediately undertake a full adjustment of productive capacity to effective demand; rather, such adjustments occur by a flexible accelerator process. Notwithstanding, a flexible degree of capacity utilisation allows firms to meet all expected peaks of demand with the current installed capacity (Ciccone 1986).

12 Such theoretical relationship is confirmed by a recent empirical paper which examines the impact of public R&D on private one on a panel of twenty-six countries (Moretti, Steinwender, and Van Reenen 2016) and for the US economy (Deleidi and Mazzucato 2019).

13 In order to increase net exports by means of an exchange rate devaluation, the Marshall-Lerner condition has to be satisfied: the sum of export and import price elasticities has to be greater than 1.

14 Yet, despite we have assumed that the exchange rate can affect the trade balance, the ability of the exchange rate to stimulate export and output growth has been questioned by several economists (see among others, Alejandro 1963; Krugman and Taylor 1978; Frenkel and Taylor 2006; Dvoskin, Feldman, and Ianni 2019).

15 In order to have an economically significant solution, the denominator of equation 10 has to be positive.

16 For an in-depth review on the notion of a fully adjusted position, see among others Serrano (1995), Cesaratto, Serrano, and Stirati (2003), Cesaratto (2015), and Freitas and Serrano (2015). Additionally, for a review of the static and local dynamic stability conditions of the SSM, see Freitas and Serrano (2015) and Lavoie (2016). Furthermore, for the stability condition of this specific model based on the investment function as in equation (5), see Pariboni (2015, Appendix A, pp. 87–89, equations A9–A11). Specifically, to have the local stability of the model, the marginal propensity to spend has to be less than one. In our model, $v\ast \left(x+d+g_{\mathrm{z}}\right)+c_{\mathrm{w}}\ast \omega \ast \left(1-t_{\mathrm{w}}-t_{\mathrm{r}\mathrm{w}}\right)+c\ast \left(1-\omega \right)\ast \left(1-t-t_{\mathrm{r}}\right)<1$.

17 In a fully adjusted position, capacity follows the trend of effective demand, $g_{\mathrm{e}}=g_{\mathrm{z}}$ and the degree of capacity utilisation is equal to the planned utilisation rate ($u=u_{\mathrm{n}}$). (Cesaratto, Serrano, and Stirati 2003, 44).

18 In a fully adjusted position where $g_{\mathrm{e}}=g_{\mathrm{z}}$, the trend growth rate generated by alternative government spending will be the same:$\mathrm{\Delta }{G}_1\frac{1}{s-v\left(d+g_{\mathrm{z}}\right)}=\mathrm{\Delta }{Y}_{G1}=\mathrm{\Delta }{Y}_{G2}=\frac{1+\gamma }{s-v\left(d+g_z\right)}\mathrm{\Delta }{G}_2.$Only a further change in the growth rate of autonomous components can generate the inequality shown in (18).

19 Here a twofold issue has to be highlighted. First, compared to the remaining variables presented in inequality (20), exogenous taxes have to decrease in order to generate a positive effect on output (see equation 10); second, for the sake of simplicity, we do not consider any affects in the change of the tax rate and of transfers coefficient.

20 For an in-depth review on the Kaldor-Verdoorn law, see among others McCombie, Pugno, and Soro (2002), Jeon and Vernengo (2008), McCombie and Spreafico (2015), Deleidi, Paternesi Meloni, and Stirati (2018) and Antenucci, Deleidi, and Paternesi Meloni (2019).

21 If the new production technique, which has been discovered by the process of innovation, is dominant for all distribution combinations compared to the old techniques, then the new technique will be adopted. That occurs since it allows firms to generate a greater rate of profit for every possible wage rate. Conversely, if the new technique is dominant, but not for all distribution combinations, it will only be adopted under certain distributive arrangements. Especially for a given profit rate, the chosen technique will have to be so to maximize the wage rate. In that case, either the new or an old production technique could be adopted.

22 The technical progress could lead to changes in the marginal propensity to consume by means of changes in the distributive shares. Such modifications could occur in both directions, leaving the change of the marginal propensity to consume undetermined. However, if the marginal propensity to consume decreases due to an increase in profit share, the stylised fact of the constancy of average propensity to consume in advanced countries (Cesaratto, Serrano, and Stirati 2003) could be explained by an increase in autonomous consumptions financed through the credit market.

23 For an in-depth discussion of the use of the rate of growth of the capital stock and the investment-output ratio, see among others Kaldor (1967), McCombie (2002) and Deleidi, Paternesi Meloni, and Stirati (2018).

24 See Nah and Lavoie (2019) for a discussion of the growth and level effects during the traverse from one to another steady state as well as the effect of the average rate of growth of output on labour productivity. As already discussed in footnote 18, the trend growth rate generated by alternative government spending shocks of the same magnitude will be the same.

Additional information

Funding

We would like to thank the European Union Horizon 2020 (DOLFINS Nr. 640772 and ISIGrowth Nr. 649186) and Innovate UK grants for financing the study.