COVID-19, firm innovation strategy and production efficiency: A stochastic frontier analysis of Caribbean firms

Mohan, Preeya; Strobl, Eric Working Paper COVID-19, firm innovation strategy and production efficiency: A stochastic frontier analysis of Caribbean firms IDB Working Paper Series, No. IDB-WP-1396 Provided in Cooperation with: Inter-American Development Bank (IDB), Washington, DC Suggested Citation: Mohan, Preeya; Strobl, Eric (2023) : COVID-19, firm innovation strategy and production efficiency: A stochastic frontier analysis of Caribbean firms, IDB Working Paper Series, No. IDB-WP-1396, Inter-American Development Bank (IDB), Washington, DC, https://doi.org/10.18235/0004775 This Version is available at: https://hdl.handle.net/10419/289953 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc-nd/3.0/igo/legalcode IDB WORKING PAPER SERIES Nº IDB-WP-1396 C C O O V V I I D D - - 1 1 9 9 , , F F i i r r m m I I n n n n o o v v a a t t i i o o n n S S t t r r a a t t e e g g y y a a n n d d P P r r o o d d u u c c t t i i o o n n E E f f f f i i c c i i e e n n c c y y : : A A S S t t o o c c h h a a s s t t i i c c F F r r o o n n t t i i e e r r A A n n a a l l y y s s i i s s o o f f C C a a r r i i b b b b e e a a n n F F i i r r m m s s Prepared for the Inter-American Development Bank by: Preeya Mohan Eric Strobl Inter-American Development Bank Institutions for Development Sector Competitiveness, Technology, and Innovation Division Compete Caribbean Partnership Facility March 2023 March 2023 C C O O V V I I D D - - 1 1 9 9 , , F F i i r r m m I I n n n n o o v v a a t t i i o o n n S S t t r r a a t t e e g g y y a a n n d d P P r r o o d d u u c c t t i i o o n n E E f f f f i i c c i i e e n n c c y y : : A A S S t t o o c c h h a a s s t t i i c c F F r r o o n n t t i i e e r r A A n n a a l l y y s s i i s s o o f f C C a a r r i i b b b b e e a a n n F F i i r r m m s s Prepared for the Inter-American Development Bank by: Preeya Mohan (Sir Arthur Lewis Institute of Social and Economic Studies, University of the West Indies, St Augustine) Eric Strobl (University of Bern) Inter-American Development Bank. This work is licensed under a Creative Commons IGO 3.0 AttributionNonCommercial-NoDerivatives (CC-IGO BY-NC-ND 3.0 IGO) license (http://creativecommons.org/licenses/by-nc-nd/3.0/igo/ legalcode) and may be reproduced with attribution to the IDB and for any non-commercial purpose, as provided below. No derivative work is allowed. Any dispute related to the use of the works of the IDB that cannot be settled amicably shall be submitted to arbitration pursuant to the UNCITRAL rules. The use of the IDB's name for any purpose other than for attribution and the use of IDB's logo shall be subject to a separate written license agreement between the IDB and the user and is not authorized as part of this CC-IGO license. Following a peer review process, and with previous written consent by the Inter-American Development Bank (IDB), a revised version of this work may also be reproduced in any academic journal, including those indexed by the American Economic Association's EconLit, provided that the IDB is credited and that the author(s) receive no income from the publication. Therefore, the restriction to receive income from such publication shall only extend to the publication's author(s). With regard to such restriction, in case of any inconsistency between the Creative Commons IGO 3.0 Attribution-NonCommercial-NoDerivatives license and these statements, the latter shall prevail. Note that the link provided above includes additional terms and conditions of the license. The opinions expressed in this publication are those of the authors and do not necessarily reflect the views of the Inter-American Development Bank, its Board of Directors, or the countries they represent. Cataloging-in-Publication data provided by the Inter-American Development Bank Felipe Herrera Library Mohan, Preeya. COVID-19, firm innovation strategy and production efficiency: a stochastic frontier analysis of caribbean firms / Preeya Mohan, Eric Strobl. p. cm. — (IDB Working Papers Series ; 1396) Includes bibliographic references. 1. Technological innovations-Caribbean Area. 2. Business enterprises-Technological innovations-Caribbean Area. 3. Industrial productivity-Caribbean Area. 4. Coronavirus infections-Economic aspects-Caribbean Area. I. Strobl, Eric, 1969- . II. Inter-American Development Bank. Competitiveness, Technology and Innovation Division. III. Title. IV. Series. IDB-WP-1396 http://www.iadb.org Copyright © 2023 Preeya Mohan Eric Strobl Prepared for the Inter-American Development Bank by: COVID-19, FIRM INNOVATION STRATEGY AND PRODUCTION EFFICIENCY A Stochastic Frontier Analysis of Caribbean Firms 1 CONTENTS ABSTRACT ..................................................................................................................................................... V 1 INTRODUCTION ................................................................................................................................ 1 2 LITERATURE REVIEW...................................................................................................................... 3 3 COVID-19, INNOVATION AND PRODUCTION EFFICIENCY AND CARIBBEAN FIRMS ......... 5 4 DATA AND METHODOLOGY ........................................................................................................... 7 4.1. Data ................................................................................................................................................................ 7 4.2. Methodology ............................................................................................................................................... 8 5 RESULTS............................................................................................................................................. 11 5.1. Production Function ................................................................................................................................... 11 5.2. Determinants of Technical Efficiency ................................................................................................... 12 6 DISCUSSION ...................................................................................................................................... 17 6.1. Policy Implications ...................................................................................................................................... 17 6.2. Limitations of the study ........................................................................................................................... 19 7 CONCLUSIONS .................................................................................................................................. 21 REFERENCES ................................................................................................................................................. 23 iii 3 COVID-19, INNOVATION AND PRODUCTION EFFICIENCY AND CARIBBEAN FIRMS The role of innovation and productivity in economic growth is poorly understood in Caribbean SIDS, mainly due to a lack of data on firms. The Compete Caribbean Partnership Facility, in collaboration with the Inter-American Development Bank and the World Bank, has made significant strides in producing internationally comparable, statistically relevant data at the firm level. This facilitates empirical analysis that can determine what drives firm performance and innovation in the region. The evidence from this data indicates that pre-COVID-19 innovation and productivity are quite low and, indeed, are acute constraints to growth and development in Caribbean SIDS. Among the first of such studies, Ruprah, Melgarejo, and Sierra (2014) highlight that the characteristics of Caribbean businesses are not typically associated with a dynamic innovative private sector since they are generally small, old, and concentrated in tourism and retail, and ownership is predominantly local. The study used the Latin American and Caribbean Enterprise Survey (LACES) and showed that Caribbean firms performed poorly over the period 2007–2010 in terms of sales growth, employment growth, and productivity, even adjusting for lower rates of growth in the region over the period. Caribbean businesses tended to be smaller (three quarters had fewer than 20 fulltime employees), older (more than 20 years in operation), and less involved in foreign trade than their small economy counterparts. Moreover, Caribbean firms were concentrated in tourism and retail, and ownership was predominantly local. Mohan, Strobl, and Watson (2016) also used the LACES dataset to identify the relationship between productivity and innovative activity at the firm level in the Caribbean. The findings show that innovative firms exhibited higher labor productivity compared to non-innovative firms, 5 and differences in firm characteristics accounted for some of the observed differences in productivity, such as size, access to public support for innovation, ownership of patents, export behavior, foreign ownership, and cooperation with other institutions for promoting innovation. However, even after allowing for these differences, the productivity mean for innovative firms was higher and there was less dispersion in productivity than for non-innovative firms. In a follow up study Mohan, Strobl, and Watson (2017) used the Productivity, Technology, Innovation (PROTEqIN) survey to explore the impact of barriers to innovation in the Caribbean. The study showed that the proportion of firms that were innovators was relatively small (26 percent of surveyed firms), but there was a larger proportion of potential innovators (59percent of surveyed firms). The study also provided empirical results showing that financing and cost, market, knowledge, and policy and regulation barriers negatively affected innovation, with the cost barrier having the largest negative impact. Moreover, potentially innovative firms experienced more stringent barriers than existing innovators, regardless of the barrier considered. There is currently no empirical research on COVID-19 related to firm productivity and innovation in Caribbean SIDS. COVID-19 has nonetheless transformed the private sector, be it in the Caribbean or internationally. It has brought with it a dramatic reduction in firm sales and a reduction in new market needs. In particular, it has been difficult for micro and small firms that lack the resources to absorb the shock, as well as for firms in retail and tourism. According to the OECD as many as 2.7 million companies in Latin America and the Caribbean are likely to close, most of them micro-enterprises, which would incur the loss of 8.5 million jobs (OECD et al. 2020). Firms have had to be innovative to survive during the COVID-19 crisis. They have had to change their operations and economic activities. The use of the internet and the adoption of digital technologies have been critical to sustaining continuity in most businesses. Businesses have launched new products and services and/or adapted to remote, teleworking, and other flexible ways of working, as well as to virtual meetings using information communications technology and digital technology (Santos, 2020). However, in the Caribbean there is a huge digital divide, notably a lack of highspeed broadband internet and a lack of appropriate digital skills. Women have also been disproportionally affected by the crisis in the region, whether at home or in the private sector, which may place female owned and managed firms at a disadvantage. There has also been a push for green growth strategies as economies seek to rebuild, given that the climate crisis in the Caribbean requires urgent attention. 6 COVID-19, FIRM INNOVATION STRATEGY AND PRODUCTION EFFICIENCY 4 DATA AND METHODOLOGY 4.1. Data Our data source is the 2021 IFPG firm-level dataset for the Caribbean collected by the Compete Caribbean Partnership Facility, a multi-donor program financed by the Inter-American Development Bank, the United Kingdom’s Foreign and Commonwealth Development Office, the Caribbean Development Bank, and the Government of Canada. This data is a representative cross-sectional enterprise survey covering 1,979 firms across 13 Caribbean countries (Barbados, Belize, Jamaica, Guyana, Suriname, Antigua-Barbuda, Dominica, Grenada, St. Kitts and Nevis, St. Lucia, St. Vincent and the Grenadines, the Bahamas, and Trinidad and Tobago) across all sectors, and conducted in 2020 during the COVID-19 pandemic.2 While the survey collected a vast array of information related to firm performance, the study here is focused on three groups of variables. The first set relates to the factors that determine a firm’s production function in order to estimate technical efficiency. In this regard, information was used to estimate technical efficiency during two phases, the pre-pandemic phase and the COVID-19 phase. For the former, respondents were asked to provide information on total sales, and the breakdown of the cost of inputs, where these were grouped into the net value of fixed assets, costs of labor, and outlay on other inputs (raw materials, energy, etc.). For the latter, where the survey was undertaken during the pandemic, respondents were asked whether sales and outlays on labor and other inputs would be negatively affected by COVID-19, and, if so, by what percentage. Using this information, an outline of expected sales, labor costs, and input expenditure during COVID-19 was drawn up. Since no specific question was raised concerning the expected effect on the net value of assets, it was assumed that this would be the same as for the previous period. The second group of variables processed from the data set were those relating to innovation. In this regard, enterprises were asked whether they had introduced any general 2 For further details of the data collection see IDB (2021). 7 innovations related to a variety of areas, including new goods or services, methods of production, logistics, information processing and communication, methods for accounting and administrative operations, business practices, work organization, and promotion. If the answer in any of these cases was affirmative, we created dummy variables to indicate as such. The respondents were also asked if any of these innovations were negatively affected by the COVID-19 crisis, and we similarly created a categorical variable to represent this. Businesses were also asked whether they had introduced, since the advent of COVID-19, any new innovation, in any of the categories listed above, and for this we also generated an indicator variable. A set of questions was also posed with regard to ‘green’ innovations in the last three years, where these were defined as innovations leading to environmental improvements intentionally or unintentionally related to reduced material consumption, reduced energy consumption, reduced CO2 footprint, less polluting or hazardous materials, reduced soil, water, noise or air pollution, or recycled waste, water, or materials. Again, businesses were asked whether such innovations were affected by COVID-19 and if similar new innovations had been introduced following the outbreak. In the same way as for general innovations, dummies were created to look at green innovations in the past, asking whether these innovations were affected by COVID-19, and if any new innovations had been introduced since the outbreak. The third body of information extracted from the Survey consists of general enterprise features that could pertain to technical efficiency, including share of the largest owner/ manager, share of foreign ownership, share of state ownership, whether the largest owner/ manager was female, and the years of experience of the manager/owner. Names and definitions of all variables used in the empirical analysis are provided in Tables 1 and 3. 4.2. Methodology In the traditional approach to productivity analysis by non-frontier models, there is the assumption that all economic agents are efficient and productivity growth takes place as a movement of the production frontier, i.e., technical change (Solow, 1957). In the event of technical inefficiency, the estimation of technical progress would be biased, and even in the absence of inefficiency, the accounting estimation of total factor productivity would be biased if individuals do not minimize cost, i.e., exhibit allocative inefficiency. The frontier approach accounts for possible inefficient behavior by measuring inefficiency as the potential increase in the observed value of production against the maximum technically achievable value defined by the production frontier (Coelli et al. 2005). To calculate a firm’s inefficiency a SFA was employed, that assumes that the form of the production function is known and therefore other parameters of the production technology need to be estimated. Importantly, SFA allows for the measurement of inefficiency and any external shocks outside the control of the firm, such as COVID-19, and which affect output level (Coelli,1996; Coelli et al., 2005; O’Donnell, Chamber, and Quiggin, 2009; Wadud, 2003). This was estimated as follows: lnYi = b 0 + b 1ln(Ki) + b 2ln(Li) + b 13ln(Mi) + b xXi + (Vi – Ui) (1) where: where: i = 1,…, N are the number of firms in the sample; Yi = sales from firm i; Ki = net value of fixed assests for firm i; Li = total expenditure on labor for firm i; Mi = total value of other inputs i; Xi = vector of country and sector of operation indicator variables; Vi = error term for firm i; Ui = a non-negative random variable for firm i, accounting for technical inefficiency in the production function. 8 COVID-19, FIRM INNOVATION STRATEGY AND PRODUCTION EFFICIENCY Crucially, the error term in (1) is broken down into two components. The first error component, Vi, is assumed to follow a symmetric distribution and is the standard error, and the other component, Ui, pinpoints the firm’s technical inefficiency and is modelled to follow an exponential distribution, therefore calculated using the mode of its distribution. One should note that the variance of U was allowed to differ by country and sector of firms, and that standard errors were clustered at the country-sector level. The concept of technical efficiency (TE), defined simply as -Ui, was used in all subsequent analysis to ease interpretation. Accordingly, if TE is equal to zero the firm is defined as being technically efficient and is at its maximum output level given the inputs used and technology available. If TE is less than zero the firm is defined as being technically inefficient. Equation (1) was estimated separately, in terms of firms’ sales and inputs for the fiscal year preceding COVID-19 (TEPC) as well as expected sales and inputs implemented in the wake of COVID-19 (TEC), providing a pre-COVID-19 and an expected COVID-19 inefficiency score for each firm. To determine whether COVID-19 induced general innovations or green innovations affecting a firm’s efficiency, the following linear model was estimated separately for TEPC: TEPC = a0 + a1INNOV + a2 GREEN + aZZ + ei (2) where: INNOV is an indicator variable of whether the firm introduced any kind of innovation in the last three years; GREEN is an indicator variable showing whether the firm introduced any green innovation in the last three years; and Z is a vector of other potential factors related to the technical efficiency of firms, as evidenced by the data, including predominantly female ownership or shareholder indicator variable (FEMALE), percentage of foreign ownership (FOREIGN), percentage of state ownership (STATE), years of experience of manager/owner (EXPERIENCE), an indicator variable for export activity (EXPORT), and percentage of ownership of largest owner (LOWNER). Equation (2) is estimated using OLS, clustering standard errors at the country-sector level. One should note that in this second stage we make the crucial assumption that the covariates included were not omitted variables correlated with the inputs in Equation (1). One should note that this may induce some inefficiency in the estimates in (1).3 Equation (2) was also estimated for the expected values owing to COVID-19, but several additional determinants related to the impact of COVID-19 were added: TEPC = a0 + a1 INNOV + a2 GREEN + a3 INNOVAFF + a4 GREENAFF + a5 INNOVCOV + a6 GREENCOV + aZ Z + ai (3) where: INNOVAFF is an indicator variable determining whether an innovation implemented in the three years prior to COVID-19 was affected by the pandemic; GREENAFF is an indicator variable to determine whether the innovation implemented in the three years prior to COVID-19 was affected by the pandemic; INNOVCOV is an indicator variable indicating whether any innovation was introduced since the outbreak of COVID-19; and GREENCOV is an indicator variable indicating whether any such innovations were ‘green’. Equation (3) is similarly estimated with OLS with standard errors clustered at the countrysector level. 3 We did experiment with estimating a model that specifies the mean of the truncated normal distribution in terms of a linear function of the inputs, but this produced abnormally high standard errors; the implication that all inputs were insignificant predictors of output is likely to be unrealistic. 9 DATA AND METHODOLOGY Finally, Equation (3) was re-estimated, but in this instance using the difference between technical efficiency before COVID-19 and the expected technical efficiency arising from COVID-19 (DTE) as the dependent variable, thus allowing one to determine factors associated with any expected changes in technical efficiency, as opposed to just levels of technical inefficiency. This means that any fixed unobservable remaining in the technical inefficiency terms can be removed by the first difference. Also, conceptually, innovation is a shift in the production function and in technical efficiency, so an innovation-augmented production function is better estimated in first difference. Given the limited nature of our data, ie., that it is cross-sectional and lacks extensive controls, it can be argued that all estimated coefficients are to be interpreted causally. 10 COVID-19, FIRM INNOVATION STRATEGY AND PRODUCTION EFFICIENCY 5 RESULTS 5.1. Production Function Summary statistics of the variables used in estimating the stochastic production function, that is, Equation (1) are provided in Table 1. The average annual sales of the firms for the fiscal year prior to COVID-19 in the sample were US$ 2.5million, but with considerable variation, hinting at large differences in firm size across sectors and countries. Examining the production function inputs, on average the costliest input was other input (inputs other than capital and labor), constituting nearly 52 percent of total input costs. In contrast, the capital stock cost was 35 percent, while the cost of labor was the cheapest, at 13 percent. During COVID-19 firms anticipated that their output would fall by 23 percent on average, while the costs of labor were anticipated to rise by over 70 percent, and the outlay on other inputs was likely to increase marginally. As noted above, because of data restrictions there was an assumption that the value of the capital stock remained constant prior to and during COVID-19. A study by Acevedo et al. (2021) using the IFPG data set supports this where it was found that approximately 2 percent of firms expected conditions to worsen related to working capital or fixed assets since the pandemic. The results from estimating the stochastic production frontier are provided in Table 2, where Column (1) provides the results for the fiscal year previous to COVID-19 and Column (2) results during COVID-19. All inputs are highly statistically significant and positive. Given that output and inputs are estimated in log form, the estimated coefficients are interpreted as elasticities. Prior to COVID-19 the elasticity of other input (other than capital and labor) was highest, at 55 percent, followed by labor at 45 percent and capital at 2 percent. The estimated elasticities for the expected values of production inputs under COVID-19 were similar to those from the previous fiscal year (other input59 percent, labor36 percent, and capital5 percent). Az-test for each input in both periods could not reject the null hypothesis that they are not statistically different. One should note that for both periods one could reject the null hypothesis of constant returns to scale. 11 The estimated technical efficiency of the production function for the fiscal year previous to COVID-19 and the expected values during COVID-19, as well as their difference, are all shown in Table 2. The average technical efficiency was –0.0324 in the year preceding COVID-19, and is expected to fall to –0.0664 due to COVID-19, that is, over 100 percent. A t-test confirmed that the fall is statistically significant. There is also considerable variation (three times the mean) in technical efficiency across firms, in the year before COVID-19 and during COVID19, as well as the difference between these two. 5.2. Determinants of Technical Efficiency Summary statistics for firm innovative activity and the explanatory variables affecting technical efficiency estimated in Equations (2) and (3) are shown in Table 3. As can be seen, around 40 percent of firms carried out some form of general innovation over the three years prior to the pandemic, and almost 50 percent underwent some sort of green innovation. Of those implementing general innovation, approximately 11 percent felt that their innovation had been affected by COVID-19, and out of those firms conducting green innovations, 43 percent considered they had been affected by the pandemic. COVID-19 induced a little over 11 percent of firms to be innovative in a general sense, and 17 percent in an environmentally friendly sense. Looking at firm characteristics, the largest owner across the sample owned about 12 percent of the firm, whereas foreign and state ownership were on average 8 and 0.1 percent, respectively. Firm managers/owners had around 21 years’ experience, and approximately 23 percent were female. A little over a third of firms did some exporting. Results of estimating Equation (2), which looks at the pre-pandemic period, are provided in Column (1) of Table 4. General innovations increased technical efficiency. On the other hand, green innovations reduced technical efficiency. A t-test of the sum of the coefficients suggests that one cannot reject that the sum of the coefficients is significantly different from zero. With regard to non-innovation related factors, only manager/owner’s experience was statistically significantly associated with technical efficiency of a firm, indicating that more experience leads to lower technical efficiency. Examining the results from Equation (3), which focuses on the period during COVID-19, one finds that there is an overall efficiency-boosting effect of general innovation, and that this effect is greater if the innovation was not affected by COVID-19 than in the case of those innovations taking place in the previous fiscal year, and, moreover, significantly so (as suggested by a z-test). However, for the firms where this innovation was said to be affected by COVID-19, the technical efficiency effect was significantly reduced. However, a t-test suggests that even for such firms, an overall positive impact of innovation on efficiency remains. In contrast, there was no significant effect from innovation undertaken as a result of COVID-19. There was no impact from green innovation introduced over the last few years, regardless of whether it was stated to have been affected by COVID-19, or in terms of green innovation introduced in response to the pandemic. In terms of the non-innovation related variables, again, only experience was a (negative) significant predictor of technical efficiency. While the impact of this was somewhat greater on predicted technical efficiency, a z-test suggests that the difference is not statistically meaningful. The final regression exercise involves using the change of technical efficiency as the regressor, controlling for the same factors as used in Equation (3). Accordingly, of the non-innovation factors included in the empirical model, only LOWNER was significant, suggesting 12 COVID-19, FIRM INNOVATION STRATEGY AND PRODUCTION EFFICIENCY Variables Definition Mean Std dev Min Max Y PC (US$ million) Sales 2.544 13.34 0.00587 490.8 K PC (US$ million) Capital Stock Value 0.475 1.902 0.000531 49.11 L PC (US$ million) Labor Cost 0.193 0.707 2.80e–05 10.82 M PC (US$ million) Cost of Other Inputs 1.317 9.404 0.000470 369.0 Y C (US$ million) Sales 1.939 11.22 0.000912 417.2 K C (US$ million) Capital Costs 0.475 1.902 0.000531 49.11 L C (US$ million) Labor Cost 0.330 1.333 3.39e–05 21.78 M C (US$ million) Cost of Other Inputs 1.247 9.408 0.000470 369.0 TE PC Technical Efficiency –0.0348 0.175 –2.792 0 TE C Technical Efficiency –0.0518 0.218 –3.571 0 ΔTE TE PC -TE C –0.0170 0.160 –2.309 0.876 Source: Authors’ compilation based on the IFPG database. TABLE 1 SUMMARY STATISTICS: PRODUCTION FUNCTION (1) (2) log(M) 0.551*** (0.0179) 0.585*** (0.0179) log(K) 0.0194*** (0.00646) 0.0449*** (0.00853) log(L) 0.454*** (0.0226) 0.360*** (0.0237) SAMPLE: PRE-COVID-19 COVID-19 Observations 1,883 1,883 Source: Authors’ compilation based on the IFPG database. Notes: (i) Country-sector clustered standard errors in parentheses, (ii)***, **, and * indicate 1, 5, and 10 percent significance levels, (iii)Country and sector dummies included but not reported, (iv) where s m is modelled as a function of country and sector dummies. TABLE 2 TECHNICAL EFFICIENCY ESTIMATION RESULTS that the larger the share of the largest owner, the greater the increase in expected technical efficiency. As for the expected technical efficiency regression in levels, one finds that general innovations undertaken in the three years prior to the pandemic and that were affected by COVID-19, reduced technical efficiency. Furthermore, the additional innovations introduced due to COVID-19 also lowered technical efficiency. If a green innovation was implemented prior to the pandemic and was affected by it, then this increased expected technical efficiency. 13 RESULTS Variables Definition Mean Std dev Min Max LOWNER % of largest owner 12.27 51.84 0 99 FOREIGN % of foreign ownership 8.482 23.79 0 100 STATE % of state ownership 0.0795 1.783 0 45 EXPERIENCE Experience of owner (years) 21.25 13.25 1 58 EXPORT Export incidence 0.364 0.481 0 1 FEMALE Female owner incidence 0.227 0.419 0 1 INNOV Innovation incidence before COVID-19 0.389 0.488 0 1 INNOVCOV Innovation incidence during COVID-19 0.119 0.324 0 1 INNOVAFF Innovation affected by COVID-19 0.415 0.693 0 4 GREEN Green innovation before COVID-19 0.497 0.500 0 1 GREENAFF Green innovation affected by COVID-19 0.437 0.496 0 1 GREENCOV Green innovation during COVID-19 0.172 0.377 0 1 Source: Authors’ compilation based on the IFPG database. TABLE 3 SUMMARY STATISTICS – TECHNICAL EFFICIENCY DETERMINANTS 14 COVID-19, FIRM INNOVATION STRATEGY AND PRODUCTION EFFICIENCY 7 CONCLUSIONS COVID-19 induced a complex global crisis in that the virus is as contagious economically as it is medically. While Caribbean countries have been exposed to frequent external shocks, in particular climatic events, the pandemic is like no other. Social distancing measures implemented by Governments to save lives brought economic activity to a near-standstill, affecting demand and supply sides of the value chain, leading to severe economic contraction, and closure of businesses. This paper has aimed to contribute to a better understanding of the COVID-19 pandemic on Caribbean firms’ performance and innovation using the IFPG firm-level dataset and a SFA framework. It has provided empirical evidence on how the persistence of innovation and the type of innovation (general and green) have possibly helped firms to mitigate the negative effects of the crisis. Caribbean firms that implemented innovations prior to and during COVID-19 demonstrated improved technical efficiency, compared to firms that did not, while firms that employed green innovations before and during the pandemic did not experience a similar boost in technical efficiency. The findings arguably have significant policy relevance for the Caribbean, as government policy aims to develop the private sector by designing and implementing appropriate policies and incentives for firms to grow and engage in innovative activity and reduce market failures. It also provides valuable information for entrepreneurs and managers when crafting innovation strategy during a crisis, such as introducing new products and processes, marketing and organizational innovations and green innovations that may improve firm performance. 21 1 REFERENCES Acevedo, M., J. Lennon, S. Pereira, and P. YañezPagans. 2021. The Impacts of the COVID-19 Pandemic on Firms in the Caribbean. Development through the Private Sector. Series TN No. 29. Adam, N. A. and G. Alarifi. 2021. Innovation Practices for Survival of Small and Medium Enterprises (SMEs) in the COVID-19 Times: The Role of External Support. Journal of Innovation and Entrepreneurship 10:15 https://doi .org/10.1186/s13731–021–00156–6 Archibugi, D., A. Filippetti, and M. Frenz. 2013. The Impact of the Economic Crisis on Innovation: Evidence from Europe. Technological Forecasting & Social Change 80(7): 1247–60. https://doi.org/10.1016/j.techfore.2013.05.005 Bernard, A., B., Stephen J. Redding and P.K. Schott. 2009. Products and Productivity. Scandinavian Journal of Economics 111(4): 681–709. Caballero, R.J. and M.L. Hammour. 1991. The Cleansing Effects of Recessions. American Economic Review 84(5): 1350–68. Chen, Y. 2008. The Driver of Green Innovation and Green Image: Green Core Competence. Journal of Business Ethics 81:531–43. Coelli, T.J. 1996a. A Guide to FRONTIER Version 4.1: A Computer Program for Stochastic Frontier Production and Cost Function Estimation. Working Paper 96/07, Centre for Efficiency and Productivity Analysis. University of New England, Armidale. Coelli, T.J., D.S..P. Rao, C.J. O’Donnell, and G.E. Battese. 2005. An Introduction to Efficiency and Productivity Analysis, 2nd ed. New York: Springer. Cucculelli, M. and V. Peruzzi. 2020. Post-Crisis Firm Survival, Business Model Changes, and Learning: Evidence from the Italian Manufacturing Industry. Small Business Economics 54: 459–74. Darendeli, I. S. and T. L. Hill. 2016. Uncovering the Complex Relationships Between Political Risk and MNE Firm Legitimacy: Insights from Libya. Journal of International Business Studies47(1): 68–92. Filippetti, A. and D. Archibugi. 2011. Innovation in Times of Crisis: National Systems of Innovation, Structure, and Demand. Research Policy 40(2): 179–92. Freeman, C. 2004. Technological Infrastructure and International Competitiveness. Industrial and Corporate Change 13(3): 541–69. Hall, B.H., P. Moncada-Paternò-Castello, S. Montresor, and A. Vezzani. 2016. Financing Constraints, R&D Investments and Innovative 23 Performances: New Empirical Evidence at the Firm Level for Europe. Economics of Innovation and New Technology 25(3): 183–96. Han, H., and Y. Qian. 2020. Did Enterprises’ Innovation Ability Increase During the COVID-19 Pandemic? Evidence From Chinese Listed Companies. Asian Economics Letters 1(3). doi.org/10.46557/001c.18072 Hausman, A. and J.J. Wesley. 2014. The Role of Innovation in Driving the Economy: Lessons from the Global Financial Crisis. Journal of Business Research 67: 2720–26. Hud, M. and K. Hussinger. 2015. The Impact of R&D Subsidies during the Crisis. Research Policy 44(10): 1844–55. IDB (Inter-American Development Bank). 2021. Innovation, Firm Performance and Gender (IFPG) Issues in Enterprises in the Caribbean Survey 2020 – Survey Description and Technical Report. --------. 2022. Finance for Firms: Options for Improving Access and Inclusion. Caribbean Economics Quarterly Volume 11 Issue 2 July 2022. https://publications .iadb.org/publications/english/document /Caribbean-Economics-Quarterly-Volume -11-Issue-2-Finance-for-Firms-Options-for -Improving-Access-and-Inclusion.pdf Krammer, Sorin M.S. 2021. Navigating the New Normal: Which Firms have Adapted Better to the COVID-19 Disruption? Technovation 110:102368. doi.org/10.1016/j.technovation .2021.102368. Lee, S. M., and S. Trimi. 2020. Convergence Innovation in the Digital Age and in the COVID-19 Pandemic Crisis. Journal of Business Research 123: 14–22. Lin, C. Y. Y., and M.Y.C Chen. 2007. Does Innovation Lead to Performance? An Empirical Study of SMEs in Taiwan. Management Research News 30(2): 115–32. Makkonen, H., M. Pohjola, R. Olkkonen, and A. Koponen. 2014. Dynamic Capabilities and Firm Performance in a Financial Crisis. Journal of Business Research 67: 2707–19. Mohan, P., E. Strobl, and P. Watson. 2016. Innovative Activity in the Caribbean: Drivers, Benefits and Obstacles. In M. Grazzi and C. Pietrobelli (eds.), Firm Innovation and Productivity in Latin America and the Caribbean: The Engine of Economic Development. Washington, DC: Palgrave Macmillan and Inter-American Development Bank. Mohan, P., E. Strobl and P. Watson. 2017. Barriers to Innovation and Firm Productivity in the Caribbean. In G.Crespi, S. Dohnert, and A.Maffioli, Exploring Firm Level Innovation and Productivity in Developing Countries: The Perspective of Caribbean Small States. Washington, DC: Inter-American Development Bank. Nemlioglu, I., and S. Mallick. 2021. Effective Innovation via Better Management of Firms: The Role of Leverage in Times of Crisis. Research Policy 50(7): 104259. Nickell, S., D. Nicolitsas, and M. Patterson. 2001. Does Doing Badly Encourage Management Innovation? Oxford Bulletin of Economics and Statistics 63(1): 5–28. OECD (Organisation for Economic Co-operation and Development). 2020. Latin American Economic Outlook 2020: Digital Transformation for Building Back Better. Paris: OECD Publishing. https://doi.org/10 .1787/e6e864fb-en. O’Cass, A., and P. Sok. 2014. The Role of Intellectual Resources, Product Innovation Capability, Reputational Resources and Marketing Capability Combinations in Firm Growth. International Small Business Journal 32(8): 996–1018. O’Donnell, C.J., R.G. Chambers, and J. Quiggin. 2009. Efficiency Analysis in the Presence of Uncertainty. Journal of Productivity Analysis 33 (1): 1–17. 24 COVID-19, FIRM INNOVATION STRATEGY AND PRODUCTION EFFICIENCY Oetzel, J., and C. H. Oh. 2021. A Storm is Brewing: Antecedents of Disaster Preparation in Risk Prone Locations. Strategic Management Journal. doi.org/10.1002/smj.3272 Oh, C. H., and J.Oetzel. 2011. Multinationals’ Response to Major Disasters: How Does Subsidiary Investment Vary in Response to the Type of Disaster and the Quality of Country Governance? Strategic Management Journal 32(6): 658–81. Olley, G. S., and A. Pakes. 1996. The Dynamics of Productivity in the Telecommunications Equipment Industry. Econometrica 64 (6): 1263–97. Oura, M. M., S.N. Zilber, and E.L. Lopes. 2016. Innovation Capacity, International Experience and Export Performance of SMEs in Brazil. International Business Review 25(4): 921–32. Paunov, C. 2012. The Global Crisis and Firm’ Investments in Innovation. Research Policy 41(1): 24–35. Rexhäuser, S. and C. Rammer. 2014. Environmental Innovations and Firm Profitability: Unmasking the Porter Hypothesis. Environmental Resource Economics 57: 145–167. Rosenbusch, N., J. Brinckmann, and A. Bausch. 2011. Is innovation Always Beneficial? A Meta-Analysis of the Relationship between Innovation and Performance in SMEs. Journal of Business Venturing 26(4): 441–57. Ruprah, I., K. A. Melgarejo, and R. Sierra. 2014. Is there a Caribbean Sclerosis? Stagnating Economic Growth in the Caribbean. Washington, DC: Inter-American Development Bank. Santos, A. M. 2020. How Has COVID-19 Accelerated Digitization and Changed Consumer Preferences? Focus on the Tourism Sector, Innovate in Tourism: From Digital Transition to Smart Destination EDP Workshop, Algarve, 30 September 2020. Solow, R. M. 1957. Technical Change and the Aggregate Production Function. Review of Economics and Statistics 39: 312–20. Song, H., Y. Yang, and Z. Tao. 2020. How Different Types of Financial Service Providers Support Small and Medium-sized Enterprises under the Impact of the COVID-19 Pandemic: From the Perspective of Expectancy Theory. Frontiers of Business Research in China 14(1): 1–27. Valerie, A. 2007. Macroeconomic Context Firms’ Long-Term Behaviour. A Study of Firms’ Investment on R & D and Machinery in Argentina during the 1990s. Desarrollo Económico 47(187): 459–85. Van Beveren, I. 2012. Total Factor Productivity Estimation: A Practical Review. Journal of Economic Surveys 26(1): 98–28. Verhees, F. J., and M.T. Meulenberg. 2004. Market Orientation, Innovativeness, Product Innovation, and Performance in Small Firms. Journal of Small Business Management 42(2): 134–54. Wadud, M.A. 2003. Technical, Allocative, and Economic Efficiency of Farms in Bangladesh: A Stochastic Frontier and DEA approach. Journal of Developing Areas 37(1): 109–26. Zouaghi, F., M. Sánchez, and M.G. Martínez. 2018. Did the Global Financial Crisis Impact Firms’ Innovation Performance? The Role of Internal and External Knowledge Capabilities in High and Low Tech Industries. Technological Forecasting and Social Change 132: 92–104. Zulu-Chisanga, S., N. Boso, O. Adeola, and P. Oghazi. 2016. Investigating the Path from Firm Innovativeness to Financial Performance: The Roles of New Product Success, Market Responsiveness, and Environment Turbulence. Journal of Small Business Strategy 26(1): 51–68. 25 REFERENCES