Abstract
Purpose
In this study, we examine whether work from home (WFH) had an impact on firm productivity during the COVID-19 period.
Design/methodology/approach
We employ a panel fixed-effect model using 79,201 firm-quarter observations in a cross-country setting of 68 countries.
Findings
First, we find that firms that employed WFH contributed to real sector growth during the pandemic due to greater capital expenditure compared to otherwise. Second, we find that WFH amenable firms turned over assets better than less WFH amenable firms.
Originality/value
To the best of our knowledge, this is the first study to examine the impact of WFH on firms’ investment and efficiency using a cross-country setting.
Keywords
Citation
Gopalakrishnan, B., Sampath, A. and Srivastava, J. (2024), "Did work from home “really” work during COVID-19?", China Accounting and Finance Review, Vol. ahead-of-print No. ahead-of-print. https://doi.org/10.1108/CAFR-09-2023-0118
Publisher
:Emerald Publishing Limited
Copyright © 2024, Balagopal Gopalakrishnan, Aravind Sampath and Jagriti Srivastava
License
Published in China Accounting and Finance Review. Published by Emerald Publishing Limited. This article is published under the Creative Commons Attribution (CC BY 4.0) licence. Anyone may reproduce, distribute, translate and create derivative works of this article (for both commercial and non-commercial purposes), subject to full attribution to the original publication and authors. The full terms of this licence may be seen at http://creativecommons.org/licences/by/4.0/legalcode
1. Introduction
The COVID-19 pandemic-induced economic crisis is unprecedented in modern history (Baker et al., 2020; Huynh, Dao, & Nguyen, 2021). The spread of the virus and the subsequent stringent lockdowns severely disrupted firms’ operations, impacting global supply chains (Brinca, Duarte, & Faria-e-Castro, 2020; Guan et al., 2020). Unlike previous economic crises, the COVID-19-induced crisis is unique in two ways: (1) it impacted firms’ assets and liabilities contemporaneously, and (2) it disrupted the primary mode of workforce interaction - face-to-face. To mitigate the crisis’s effects, several firms moved their workforce remotely to continue activity and production. This distinctive crisis and subsequent firm reaction present us with an opportunity to investigate whether firms’ ability to pivot to remote working helped during the crisis period. In this study, we examine whether amenability to remote working impacted firm-level activity during the COVID-19 pandemic.
COVID-19 created a supply shock for “non-essential” industries because of the lockdowns, quarantines, and stay-at-home orders to curtail the spread of the virus (Dingel & Neiman, 2020; Koren & Pető, 2020). Additionally, global demand, especially for discretionary and durable products, has reduced due to the pandemic-induced disruptions (Guerrieri, Lorenzoni, Straub, & Werning, 2022). Del Rio-Chanona, Mealy, Pichler, Lafond, and Farmer (2020) found that the pandemic adversely impacted sectors less amenable to remote working like entertainment and air transportation. Overall, early evidence indicates that COVID-19 resulted in reduced output for the industries that faced an immediate shortfall in supply and demand. Firms that were unable to continue operations had disrupted investment cycles and supplies. Therefore, we expect firms that have adapted via remote working to be better positioned to mitigate the risks of pandemic-induced disruptions.
Firms amenable to working remotely are more likely to invest in future capacity due to their ability to carry out operations even during the pandemic. In the previous major economic crisis, the Global Financial Crisis (GFC), workplace flexibility did not impact firms’ decisions, as the GFC did not directly impact daily operations, especially face-to-face work. However, remote work played a central role during the pandemic-induced crisis in the field (Barry, Campello, Graham, & Ma, 2022). Anecdotal evidence also shows that the remote working ability of firms impacts their investment flexibility. Given this context, we expect WFH amenable firms to have better investment flexibility to increase their capital expenditure further to exploit the pandemic-induced growth opportunities (Barry et al., 2022). In this context, Bai, Brynjolfsson, Jin, Steffen, and Wan (2021) show that firms with remote working ability retained a higher share of capital investments during the pandemic period. To illustrate this, in Figure 1, we show investments in capital expenditure pre-COVID-19 and during the COVID-19 shock period. The trends in Figure 1 indicate that teleworkable industries have increased their investments in capital expenditure compared with less teleworkable industries during the COVID-19 shock period.
The defining characteristic of the COVID-19 crisis was that globally governments announced lockdowns, quarantines and stay-at-home orders to reduce the spread of the virus. These restrictions adversely impacted the firms’ operations, resulting in reduced sales. Thus, firms resorted to increased usage of WFH operations. Bai et al. (2021) show that the remote working firms have higher resilience during the pandemic. These firms have higher sales, net income, and stock returns than less remote working firms during COVID-19. As the firms amenable to remote working can continue their operations during the pandemic period, we expect such firms to have a higher asset turnover during the COVID-19 shock period than the less amenable firms. To illustrate this, we show in Figure 2 firms’ asset turnover in the pre-COVID-19 and COVID-19 shock periods. The trends shown in Figure 2 indicate that firms in industries amenable to remote working have a higher asset turnover compared with less amenable industries during the COVID-19 pandemic. Taken together, it is likely that WFH - amenable firms increased their firm-level activity during the pandemic period.
In this paper, we investigate the impact of remote working amenability on firm activity. We use a sample of 79,201 observations from 68 countries to examine the impact of remote working ability on firm-level activity. Using a difference-in-difference model, we find that firms in industries amenable to remote working increase their capital expenditure and asset turnover during the pandemic. With a large cross-country sample and interactive fixed effects model, the identification strategy provides external validity for our results. The results are robust to various alternative specifications and definitions of industries' remote working abilities.
To the best of our knowledge, this is the first study to examine the impact of remote working amenability on the activity level of firms, and our contribution is as follows. First, we find that remote working firms invest more in capital expenditure during the pandemic period. These firms invest more in future growth opportunities as they continue their operations during the pandemic. Second, the WFH amenable firms have higher asset turnover during COVID-19, indicating higher resilience and capacity utilisation by the WFH amenable firms during the pandemic. Third, we find that the remote working firms in developed economies exhibit higher capital expenditure and asset turnover than those in emerging economies. This result emphasises the impact of the digital divide between the developed and emerging economies.
Additionally, we also focus on the ability of such firms to increase capital expenditure and asset turnover by remote working firms during the COVID-19 period. Specifically, we show that high WFH amenable firms have higher cash to invest in operations during the pandemic period. Cash holdings help firms reduce the adverse impact of the pandemic and invest more in future opportunities (Fahlenbrach, Rageth, & Stulz, 2021). Furthermore, these firms also have higher debt and tangibility during the pandemic. These findings suggest that WFH - amenable firms become opportunistic during COVID-19 and raise more debt to invest in future growth as they continue their operations. The increased tangibility suggests high debt raising capacity during the pandemic period. Our results support the findings of Cherry, Jiang, Matvos, Piskorski, and Seru (2021) and Gopalakrishnan, Jacob, and Mohapatra (2022).
Our findings also add to the emerging literature of COVID-19 impact on firms, and corporate finance literature. Ding, Levine, Lin, and Xie (2021) showed that firms less exposed to global supply chains and more engaged in corporate social responsibility activities were less adversely impacted by the pandemic. Barry et al. (2022) show the impact of corporate flexibility, in terms of, workforce flexibility, investment flexibility and financial flexibility during the pandemic period. This study shows that these factors play a crucial role in business planning and responses to the crisis. Bai et al. (2021) show that the firms amenable to WFH show higher resilience in terms of sales, income and returns during COVID-19. Bloom, Fletcher, and Yeh (2021) show heterogeneity in how COVID-19 impacts firms. Offline businesses were more adversely impacted during the pandemic period. Regarding research evidence in corporate finance, previous studies show that the announcement of capital expenditure and the quality of capital expenditure impact the market’s reaction to the decision (Chung, Wright, & Charoenwong, 1998; McConnell & Muscarella, 1985). Furthermore, evidence also indicates that creditors impose capital expenditure restrictions when the credit quality of the borrower decreases. It further results in reduced firm investment (González, 2016; Nini, Smith, & Sufi, 2009). Other results also suggest that asset turnover positively impacts the profitability (Alarussi & Alhaderi, 2018; Dickinson, 2011).
In continuation of these results, our study provides imperative evidence that firms can cope with the challenges posed by the COVID-19 shock, and specifically, WFH amenable firms have higher firm-level activity during the crisis period. The rest of the paper is organised as follows. The next section reviews relevant literature and builds on the hypotheses tested in our study. The subsequent section describes the data and the research design employed in our study. Next, we discuss the key findings of our study. In the final section, we conclude with potential insights and learnings for investors and policymakers.
2. Review of relevant literature and hypothesis development
2.1 Work from home and capital expenditure
Even before the breakout of COVID-19, many firms adopted WFH. Bloom, Liang, Roberts, and Ying Jenny (2015) show that firms adopting WFH practices show higher productivity and improved performance. The study shows that adopting modern technology techniques helps improve firm performance. Innovative management practices are essential for firm productivity. Klotz (2016) also show that WFH practices result in increased firm productivity. The COVID-19 induced restrictions adversely impacted firms in general. This resulted in increased volatility of operational revenues and decreased recovery (Ge, Huang, Wang, Jiang, & Liu, 2023). Further, it had a supply chain spillover. Ge et al. (2023) show that WFH increases the firm resilience by reducing operational revenue volatility and supply chain disruptions. The WFH amenability also helps small firms survive during the pandemic. Zhang, Gerlowski, and Acs (2022) show that digital resilience helps small firms reduce the probability of default, operational disruptions, and better cash flow position. The study demonstrates WFH adoption as “creative destruction.” It also helps reduce COVID-19 distress. WFH mitigates the COVID-19-induced shock by increasing employees' probability of performing their jobs (Alipour, Fadinger, & Schymik, 2021). In addition, WFH practice is positively associated with success during the pandemic period (Kagerl & Starzetz, 2023). Given that the pandemic-induced uncertainty provides disproportionate growth opportunities to the firms that are amenable to working from home, we posit that:
Firms that are more amenable to remote working have higher capital investment intensity than the less amenable firms during COVID-19.
2.2 Asset turnover and work from home
In recent times, ample evidence exists related to WFH amenability and its impact on individuals and firms since the inception of COVID-19. Previous studies show that firms' digital resilience helps improve performance. For instance, Fairlie and Fossen (2022) show that firms suffered from sales losses during the pandemic period. These losses were the largest for firms majorly impacted by the stringent lockdowns imposed during the pandemic to curtail the virus’s spread. However, the firms that could shift to online mode increased sales even during the pandemic. The study shows that there is a shift from an in-store business to a digital business during the pandemic. In addition, the number of firms that practised digital interaction has increased more compared with traditional interaction (Gavin, Harrison, Plotkin, Spillecke, & Stanley, 2020). Additionally, the firms amenable to WFH were able to adapt to the pandemic-induced changes. Bai et al. (2021) showed that firms with high WFH amenability in the pre-pandemic period performed better during the pandemic. Barrero, Bloom, Davis, and Meyer (2021) and Bloom et al. (2021) show that the firms with a stronger WFH amenability continue to grow even after the peak crisis period. Therefore, we posit that more WFH amenable firms generate higher sales from the assets than less WFH amenable firms during COVID-19.
Firms more amenable to remote working have higher asset turnover than the less amenable firms during COVID-19.
2.3 Work from home and the digital divide
Debates on the global digital divide [1] predate COVID-19 pandemic. The United Nations Conference on Trade and Development (UNCTAD), in its Digital Economy Report 2019, impresses on the impact of digital technologies on transforming economic and social activities. Simultaneously, the report warns that “widening digital divides threaten to leave developing countries, and especially least developed countries, further behind”. Dingel and Neiman (2020), in their seminal work on working from home clearly articulate the existence of the digital divide – emerging lower-income economies have lower teleworkable jobs. Chiou and Tucker (2020) show that even within a high income economy such as the USA, there is a digital divide between high- and low income households.
On the one hand, the aforementioned anecdotal evidence points to the presence of a digital divide, especially between advanced and developing economies. On the other hand, studies such as Bloom et al. (2015), Klotz (2016), Bai et al. (2021), Barry et al. (2022) clearly point out the positive benefits of work from home amenability, and its impact on various measures of firm performance. During COVID-19, the response by all governments regardless of income level was similar – lock downs, and as a result, working from home. Combining the two perspectives of the digital divide’s presence and the impact of work from home on firm activity and performance, we find this an interesting setting to study whether the digital divide affects the impact of remote working on firm activity. Therefore, we posit that firms domiciled in advanced economies with better access to digital resources (as evidenced by income level/developmental status), relatively fared better in generating better Capex, and turning over assets than firms domiciled in emerging economies.
Work from home amenable firms domiciled in advanced economies have higher capex intensity and asset turnover than WFH amenable firms domiciled in developing economies.
3. Data and methodology
3.1 Data description and summary
We employ a quarterly panel of 79,201 observations from 68 countries [2]. Our sample period starts from January 2017 to December 2020. We define the COVID-19 shock period as Q2′2020 to Q4′2020. We obtain all firm-level variables from Thomson Reuters Eikon and exclude all financial firms from our study. In Table 1, we provide a detailed description of all the variables used in the study. We match all the WFH amenability measures of industries to our sample based on the two-digit North American Industry Classification System (NAICS) code. We winsorize all the financial variables at 1st and 99th percentiles to deal with outliers.
We define Capital Expenditure as the ratio of CapEx to assets and Asset turnover as the ratio of sales to assets. CapEx equals the expenditure incurred for factories, equipment and intangible assets with a useful life of more than a year.
Our primary explanatory variable is Xj × COVID-19, where X represents the WFH variables. Our choice of WFH variables stems from the nature of work and activity level of an occupation in an industry. Therefore, we use Remote working, Remote working wage, Team interaction, Customer interaction, Physical presence and Face2Face interaction as proxies for WFH amenability.
Remote working and Remote working wage are based on Dingel and Neiman (2020). Remote working equals the industry-wise proportion of remote working jobs. Remote working wage equals the industry-wise wage proportion of remote working jobs. These measures are based on two surveys collected by the O*NET database for 1,000 occupations in the US. These surveys focus on the factors that influence the nature of work and general type of job behaviour. In addition to this, we use High remote working and High remote working wage. These dummy variables equal 1 for the above-median values of Remote working and Remote working wage, respectively, and 0 otherwise.
Further, we use Team interaction, Customer interaction and Physical presence based on Koren and Pető (2020). These measures are based on the communication intensity required in a job. Team interaction equals the extent of coordinating the work and guiding subordinates. It is based on the internal communication required with co-workers. Customer interaction equals the extent of the requirement of establishing and maintaining interpersonal relationships with the customers. It is based on the external communication with the customers. An industry is classified as less amenable to WFH if it requires high direct customer interaction. Physical presence is based on the extent of the requirement of repairing and maintaining electronic and mechanical equipment. We also use High team interaction, High customer interaction and High physical presence based on the median values above-mentioned three measures. These dummy variables equal 1 for the above median values of Team interaction, Customer interaction and Physical presence and 0
otherwise.
Last, we also use Face2Face interaction as a proxy for WFH amenability of industries. We use this measure based on Avdiu and Nayyar (2020). Face2Face interaction is based on the extent of the requirement of working directly with others or influencing others. Again, we use high Face2Face interaction based on the median value of Face2Face interaction. It equals 1 for the above-median value of Face2Face interaction and 0 otherwise. We match all the WFH amenability measures to our sample based on the two-digit NAICS code.
We use Liquidity, Leverage, Profitability and Size as firm-level control variables. Liquidity equals cash and cash equivalents scaled by the total assets of the firm. Leverage and Profitability equal the debt-to-equity ratio and Earnings before interest, tax, depreciation and amortisation (EBITDA) scaled by total assets, respectively. We define Size as the logarithm of the total assets of the firms. Table 1 describes all the variables used in the study. We winsorize all the financial variables employed in our study at 1st and 99th percentiles to deal with outliers.
In Table 1, we also show the summary statistics. The average asset turnover is 0.26, which indicates that the firms in our sample are good at generating revenue from their assets. The average capital expenditure is 1.17% of the assets. The average remote working score and face2face interaction score are 0.33 and 1.05, respectively. The median value of remote working variables and high face2face interaction is 1, indicating that half of the sample firms are more amenable to WFH. However, according to the communication intensity variables of WFH amenability, 75 percentile of the firms in our sample are less amenable to WFH. The average size of firms in our sample is 13.92. The average profitability of firms in our sample is 0.02.
3.2 Empirical methodology
We use a difference-in-differences (DiD) method to study the impact of teleworkability of industries on firms’ activities. We employ the following empirical estimation model:
We control for firm-level time-invariant heterogeneity denoted by δi. Furthermore, we also control for time-variant changes at the country-year-quarter level represented by γcyq. These interactive fixed effects control for any unobserved time-variant changes at the country-year-quarter level besides the time and country-level changes in isolation [3]. The saturated model helps us isolate the impact of industries’ teleworkability on firms’ activities during COVID-19 shock period and also improves the identification strategy employed in our study [4]. Robust standard errors, which control for heteroscedasticity, are clustered at the firm-level to control for autocorrelation in the error structure.
3.3 Parallel trends
Figure 1 shows the parallel trends of capital expenditure in the pre-COVID-19 and COVID-19 periods. We document increased capital expenditure for firms in industries that are more amenable to WFH. For instance, firms in industries requiring high face2face interaction have lower capital expenditure during the COVID-19 period than those in industries requiring low face2face interaction.
Figure 2 shows the parallel trends of the asset turnover ratio in the pre-COVID-19 and COVID-19 periods. The parallel trends are based on the dummy variables used for WFH variables. The parallel trends are based on high remote working, high remote working wage, high team interaction, high physical presence and High Face2Face interaction. We find that asset turnover has increased for firms in the teleworkable industries during the COVID-19 period. For instance, firms in industries that are more remote working have higher asset turnover relative to firms in industries that are less remote working.
4. Results and discussion
We show the results related to Equation (1) in this section. The results related to asset turnover and capital expenditure are presented in Table 2. We repeat the analysis with continuous measures of amenability. These results are shown in Table 3. Furthermore, we divide our sample into firms in advanced economies and emerging economies based on the IMF classification. We show the subsample analysis results in Table 4.
4.1 WFH amenability and activity
In Table 2, we illustrate the results related to the impact of COVID-19 on capital expenditure (see columns (1)-(5)). Our results indicate an increase in firms' capital expenditure in industries more amenable to WFH. Table 2 shows that capital expenditure increased by 0.176% during the COVID-19 shock period for firms in industries amenable to remote working. Such an increase during the pandemic period is about 15% of the average capital expenditure for firms in our sample. Moreover, our results show that it declined by 0.124 and 0.072% for firms in industries that require high customer interaction and high face2face interaction. Our results are robust and consistent across countries, suggesting WFH amenability’s global impact on real sector growth. Firms with the benefits of amenable operations can likely continue expanding to capture growth opportunities during uncertain times. The increase in investment intensity by the remote working firms is imperative to improve the long-term growth prospects during the pandemic (Curran, 2021). These results support hypothesis 1.
Longer-term investment will be driven by trends such as supply chain diversification or accelerated automation in the service sector as workforce age.
Our results suggest that firms in teleworkable industries have substantially increased their investment intensity. The social distancing norms imposed during COVID-19 have shifted the focus to remote working. Accordingly, firms more amenable to WFH increase their capital expenditure to increase their income generating capacity.
Columns (6)-(10) in Table 2 show the results related to the impact of WFH amenability of industries on the asset turnover of firms. Our results suggest that asset turnover significantly increased for firms in industries more amenable to WFH. For instance, asset turnover increased by 0.013 units for firms in industries amenable to flexible operations in terms of jobs performed at remote locations. Furthermore, our results show that asset turnover declined by 0.015 units and 0.011 units for firms in industries that require high customer interaction and high face2face interaction (less amenable to remote working) during COVID-19 shock period. Our results are in favour of hypothesis 2.
The possible explanation for our results is the shift to a remote working culture caused by COVID-19-induced shock (Brynjolfsson et al., 2020). Firms operating remotely can increase their sales disproportionately (e.g. technology). This indicates higher capacity utilisation by the remote working firms during the COVID-19 shock period. Conversely, firms in industries that are less amenable to WFH are adversely impacted by the pandemic and consequently face a decline in asset turnover.
We also repeat our analysis using continuous measures of WFH amenability (see Table 3). Panel A in Table 3 shows the results with capital expenditure as the dependent variable, and panel B shows the results with asset turnover as the dependent variable. The results suggest that asset turnover and capital expenditure increase for firms in teleworkable industries during the COVID-19 shock period. On the contrary, it decreases for firms in industries less amenable to WFH. For instance, a one-unit increase in the face2face interaction results is 0.565% points (1.050 × 0.539) decrease in capital expenditure relative to its mean. Similarly, a one-unit increase in the face2face interaction results in a 0.084 unit (1.050 × 0.080) decrease in asset turnover relative to its mean. Our results are consistent with the findings shown in Table 2.
4.2 Do developed market firms fare better than emerging market firms?
The teleworkability of industries likely has a pronounced impact on firms in developed economies as these firms enjoy higher productivity and growth benefits. Therefore, we test whether the remote working amenability of firms in developed economies has a prominent impact on capital expenditure and asset turnover compared with firms in emerging economies during COVID-19. We divide the sample into developed or emerging economies based on the classification provided by the International Monetary Fund (IMF).
We repeat our baseline estimation for subsamples based on firms domiciled in developed and emerging economies. Table 4, panel A shows the results with capital expenditure as the dependent variable, and panel B shows the results with asset turnover as the dependent variable. Our results show that highly teleworkable firms in developed economies have higher capital expenditure and asset turnover during the COVID-19 shock period. For instance, during the pandemic, capital expenditure and asset turnover increased by 0.207% and 0.017 units for highly remote working firms in developed economies. These results support hypothesis 3.
One possible explanation is the vast digital divide in emerging markets that may constrain teleworkability in general. Our findings complement the findings of Dingel and Neiman (2020) that the number of jobs that can be performed remotely is significantly lower in emerging market economies than in developed economies. Hence, it is unsurprising to see higher activity levels in the developed economies.
4.3 Drivers of investment behaviour
We conduct several tests to analyse the drivers of investment behaviour of firms during the COVID-19 period. Fisher (1933) show that there exists a collateral channel through which the assets impact the natural activities of a firm. The reduced collateral capacity of firms reduces the debt taking capacity of firms, which in turn, reduces the investment capacity of firms (Bernanke & Gertler, 1990; Gan, 2007; Kiyotaki & Moore, 1997). In line with this argument, we check whether WFH amenable firms have higher collateral and debt-taking capacity during the COVID-19 period. We show the results in Table A2. Our results suggest that the remote working firms have higher tangibility, representing firms' collateral capacity during COVID-19. The remote working firms exhibit 0.2% higher tangibility than the less remote working firms during the pandemic. Furthermore, we also find that WFH - amenable firms have increased debt during the pandemic. These results are shown in Table A3, suggesting that such firms with high debt capacity invest more in future growth opportunities during the COVID-19 period. The remote working firms exhibited 0.3% higher debt than the less remote working firms during the pandemic. In addition, our results also show that WFH amenable firms have higher cash holdings during the pandemic period to invest in the operations (please see Table A4). Such firms show 0.2% higher cash than the less WFH amenable firms during the pandemic period. We show that the firms’ investment behaviour is mainly driven by the tangibility, debt, and liquidity during the COVID-19 period. Our results align with the findings of Huang and Mazouz (2018) and Gan (2007).
4.4 Robustness tests
We conduct a set of robustness tests to check the validity of our results. First, we conduct a robustness test by introducing a change in the treatment window. As COVID-19 had a substantial impact in some of the countries in early 2020, we define COVID-19 period starting from Q1′2020 to Q4′2020 as 1 and 0 otherwise. We repeat our baseline estimation considering the early impact of COVID-19. Table A5 shows the robustness test results. We find consistent results with the findings shown in Table 2.
Second, we conduct a propensity score matching (PSM) analysis. It is likely that there exist differences in the firm’s characteristics in the groups divided based on the remote working ability. To illustrate this, we show the mean differences in the firm characteristics in Table A6. The top panel shows the mean differences before conducting PSM. There exist significant differences in the means of leverage and size. To address this concern, we run PSM in two steps. In the first step, we run a logistic regression and use the estimated propensity scores to match the treated and control groups. In the second step, we use the matched sample of treated and control groups, which are otherwise similar but only differ in firm-level activities, and re-estimate Equation (1) [5]. We show the propensity score matching results in Table A7. Additionally, the mean differences shown in the bottom panel of Table A6 also illustrate our results. We report that the results are consistent with our baseline findings. Last, we re-estimate the baseline equation by clustering the standard errors at the industry-level. As the firm characteristics tend to be correlated with an industry, there is potential for our estimates to be biased. Therefore, we cluster the standard errors at the industry-level and show the results in Table A8. We find that our results are consistent with the results shown in Table 2.
5. Conclusion
In this article, we investigate the impact of WFH amenability of firms and their activity during the COVID-19 pandemic and document two important findings. First, firms more amenable to WFH contribute to real sector growth, as we document that their capital expenditure is significantly higher during this period compared to firms that are less amenable to WFH. Second, firms more amenable to WFH, turnover assets more than firms less amenable to WFH. Our results are consistent and robust to (1) model specifications and (2) alternate variable specifications. While work from home might be a transitory phenomenon, the activity levels of firms that were more adaptable to flexible modes of operations reveal the benefits of operational flexibility. The insights from the study help to reinforce the benefits of amenable operations in mitigating the adverse consequences of a crisis.
Figures
Variable definitions, data sources and summary statistics
| Variable | Definition and construction | Data source | Observations | Mean | SD | Median | Min | Max |
|---|---|---|---|---|---|---|---|---|
| Asset turnover | Ratio of sales to total assets of the firm | Thomson Reuters | 79201 | 0.262 | 0.177 | 0.223 | 0.017 | 0.983 |
| Capital expenditure (%) | Expenditures for factories, equipment, software development costs and intangible assets that have a useful life of more than one year as percentage of total assets | Thomson Reuters | 53497 | 1.176 | 1.284 | 0.788 | 0.000 | 7.291 |
| Remote working | A measure based on the industry-wise proportion of remote working jobs | Dingel and Neiman (2020) | 79201 | 0.332 | 0.228 | 0.277 | 0.018 | 0.930 |
| Remote working wage | A measure based on industry-wise wage proportion of remote working jobs | Dingel and Neiman (2020) | 79201 | 0.423 | 0.237 | 0.399 | 0.038 | 0.959 |
| Team interaction | A measure based on the internal communication with co-workers | Koren and Pető (2020) | 78803 | 22.619 | 11.080 | 18.000 | 5.000 | 50.000 |
| Customer interaction | A measure based on the external communication with customers | Koren and Pető (2020) | 78803 | 18.604 | 19.886 | 8.000 | 3.000 | 90.000 |
| Face2Face interaction | A measure based on the extent of personal relationships required or working directly with public in an occupation | Avdiu and Nayyar (2020) | 79201 | 1.050 | 0.195 | 0.968 | 0.832 | 1.719 |
| High remote working | A dummy variable that equals 1 for above-median Remote working score and 0 otherwise | Dingel and Neiman (2020) | 79201 | 0.536 | 0.499 | 1.000 | 0.000 | 1.000 |
| High remote working wage | A dummy variable that equals 1 for above-median Remote working wage score and 0 otherwise | Dingel and Neiman (2020) | 79201 | 0.517 | 0.500 | 1.000 | 0.000 | 1.000 |
| High team interaction | A dummy variable that equals 1 for above-median Team interaction score and 0 otherwise | Koren and Pető (2020) | 78803 | 0.500 | 0.500 | 0.000 | 0.000 | 1.000 |
| High customer interaction | A dummy variable that equals 1 for above-median Customer interaction score and 0 otherwise | Koren and Pető (2020) | 78803 | 0.483 | 0.500 | 0.000 | 0.000 | 1.000 |
| High Face2Face interaction | A dummy variable that equals 1 for above-median Face2Face interaction and 0 otherwise | Avdiu and Nayyar (2020) | 79201 | 0.558 | 0.497 | 1.000 | 0.000 | 1.000 |
| Liquidity | Cash and cash equivalents scaled by total assets | Thomson Reuters | 79201 | 0.146 | 0.146 | 0.101 | 0.000 | 0.675 |
| Leverage | Debt-to-Equity ratio of the firm | Thomson Reuters | 79201 | 0.752 | 1.335 | 0.406 | −3.151 | 8.245 |
| Profitability | Earnings before interest, tax, depreciation and amortisation (EBITDA) scaled by total assets of the firm | Thomson Reuters | 79201 | 0.024 | 0.027 | 0.023 | −0.088 | 0.111 |
| Size | Logarithm of total assets of the firm (USD) | Thomson Reuters | 79201 | 13.291 | 2.019 | 13.223 | 6.891 | 17.812 |
| Δ Tangibility | Change in tangibility scaled by total assets | Thomson Reuters | 75602 | 0.004 | 0.025 | 0.001 | −0.072 | 0.148 |
| Δ Debt | Change in debt scaled by total assets | Thomson Reuters | 75877 | 0.005 | 0.044 | 0.000 | −0.129 | 0.220 |
| Δ Cash | Change in cash scaled by total assets | Thomson Reuters | 76871 | 0.004 | 0.047 | 0.0000 | −0.142 | 0.213 |
Source(s): Authors’ computations
WFH amenability and activity
| Capital expenditure | Asset turnover | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| (1) | (2) | (3) | (4) | (5) | (6) | (7) | (8) | (9) | (10) | |
| High remote working × COVID-19 | 0.176*** | 0.013*** | ||||||||
| (0.032) | (0.002) | |||||||||
| High remote working wage × COVID-19 | 0.147*** | 0.012*** | ||||||||
| (0.032) | (0.002) | |||||||||
| High team interaction × COVID-19 | 0.144*** | 0.013*** | ||||||||
| (0.033) | (0.002) | |||||||||
| High customer interaction × COVID-19 | −0.124*** | −0.015*** | ||||||||
| (0.031) | (0.002) | |||||||||
| High Face2Face interaction × COVID-19 | −0.072** | −0.011*** | ||||||||
| (0.031) | (0.002) | |||||||||
| Liquidity | 0.432*** | 0.431*** | 0.418*** | 0.410*** | 0.418*** | −0.085*** | −0.085*** | −0.086*** | −0.087*** | −0.086*** |
| (0.114) | (0.114) | (0.115) | (0.115) | (0.114) | (0.009) | (0.009) | (0.009) | (0.009) | (0.009) | |
| Leverage | −0.022*** | −0.023*** | −0.024*** | −0.023*** | −0.023*** | −0.002** | −0.002** | −0.002** | −0.002** | −0.002** |
| (0.007) | (0.007) | (0.007) | (0.007) | (0.007) | (0.001) | (0.001) | (0.001) | (0.001) | (0.001) | |
| Profitability | 1.109*** | 1.126*** | 1.149*** | 1.195*** | 1.205*** | 0.129*** | 0.130*** | 0.136*** | 0.139*** | 0.135*** |
| (0.388) | (0.388) | (0.391) | (0.391) | (0.388) | (0.028) | (0.028) | (0.028) | (0.028) | (0.028) | |
| Size | 0.013 | 0.013 | 0.012 | 0.012 | 0.013 | −0.003*** | −0.003*** | −0.003*** | −0.003*** | −0.003*** |
| (0.010) | (0.010) | (0.010) | (0.010) | (0.010) | (0.000) | (0.000) | (0.000) | (0.000) | (0.000) | |
| Constant | 0.932*** | 0.936*** | 0.948*** | 0.957*** | 0.946*** | 0.315*** | 0.315*** | 0.314*** | 0.315*** | 0.316*** |
| (0.134) | (0.134) | (0.135) | (0.135) | (0.134) | (0.006) | (0.006) | (0.006) | (0.006) | (0.006) | |
| Firm fixed effects | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| Country-year-quarter fixed effects | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| Observations | 53,407 | 53,407 | 53,137 | 53,137 | 53,407 | 79,201 | 79,201 | 78,876 | 78,876 | 79,201 |
| Adjusted R2 | 0.521 | 0.521 | 0.521 | 0.521 | 0.521 | 0.915 | 0.915 | 0.915 | 0.915 | 0.915 |
Note(s): The dependent variable in columns (1)-(5) is Capital expenditure and that in columns (6)-(10) is Asset turnover. The description of all variables is presented in Table 1. The standard errors are shown in parenthesis, which are clustered at the firm level. ***, **, *denotes significance level at 1, 5 and 10% respectively
Source(s): Authors’ computations
WFH amenability and activity: Estimations with continuous measure of amenability
| (1) | (2) | (3) | (4) | (5) | |
|---|---|---|---|---|---|
| Panel A – capital expenditure | |||||
| Remote working × COVID-19 | 0.316*** (0.070) | ||||
| Remote working wage × COVID-19 | 0.332*** (0.068) | ||||
| Team interaction × COVID-19 | 0.005*** (0.002) | ||||
| Customer interaction × COVID-19 | −0.005*** (0.001) | ||||
| Face2Face interaction × COVID-19 | −0.539*** (0.075) | ||||
| Firm-level controls | Yes | Yes | Yes | Yes | Yes |
| Firm fixed effects | Yes | Yes | Yes | Yes | Yes |
| Country-year-quarter fixed effects | Yes | Yes | Yes | Yes | Yes |
| Observations | 53,407 | 53,407 | 53,137 | 53,137 | 53,407 |
| Adjusted R2 | 0.522 | 0.522 | 0.522 | 0.522 | 0.522 |
| Panel B – asset turnover | |||||
| Remote working × COVID-19 | 0.038*** | ||||
| (0.005) | |||||
| Remote working wage × COVID-19 | 0.038*** | ||||
| (0.005) | |||||
| Team interaction × COVID-19 | 0.001*** | ||||
| (0.000) | |||||
| Customer interaction × COVID-19 | −0.001*** | ||||
| (0.000) | |||||
| Face2Face interaction × COVID-19 | −0.080*** | ||||
| (0.007) | |||||
| Firm-level controls | Yes | Yes | Yes | Yes | Yes |
| Firm fixed effects | Yes | Yes | Yes | Yes | Yes |
| Country-year-quarter fixed effects | Yes | Yes | Yes | Yes | Yes |
| Observations | 79,201 | 79,201 | 78,876 | 78,876 | 79,201 |
| Adjusted R2 | 0.915 | 0.915 | 0.915 | 0.915 | 0.915 |
Note(s): The dependent variable in Panel A is Capital expenditure and that in Panel B is Asset turnover. The description of all variables is presented in Table 1. The standard errors are shown in parenthesis, which are clustered at the firm level. ***, **, *denotes significance level at 1, 5 and 10% respectively
Source(s): Authors’ computations
WFH amenability and activity: Developed vs emerging market economies
| Developed economies | Emerging economies | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| (1) | (2) | (3) | (4) | (5) | (6) | (7) | (8) | (9) | (10) | |
| Panel A – capital expenditure | ||||||||||
| High remote working × COVID-19 | 0.207*** | 0.064 | ||||||||
| (0.035) | (0.072) | |||||||||
| High remote working wage × COVID-19 | 0.171*** | 0.044 | ||||||||
| (0.034) | (0.071) | |||||||||
| High team interaction × COVID-19 | 0.116*** | 0.167** | ||||||||
| (0.035) | (0.074) | |||||||||
| High customer interaction × COVID-19 | −0.132*** | −0.114* | ||||||||
| (0.034) | (0.069) | |||||||||
| High Face2Face interaction × COVID-19 | −0.134*** | 0.049 | ||||||||
| (0.033) | (0.071) | |||||||||
| Firm-level controls | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| Firm fixed effects | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| Country-year-quarter fixed effects | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| Observations | 32,355 | 32,355 | 32,126 | 32,126 | 32,355 | 19,136 | 19,136 | 19,114 | 19,114 | 19,136 |
| Adjusted R2 | 0.552 | 0.552 | 0.552 | 0.552 | 0.552 | 0.495 | 0.495 | 0.495 | 0.495 | 0.495 |
| Panel B – asset turnover | ||||||||||
| High remote working × COVID-19 | 0.017*** | 0.002 | ||||||||
| (0.003) | (0.004) | |||||||||
| High remote working wage × COVID-19 | 0.016*** | 0.001 | ||||||||
| (0.003) | (0.004) | |||||||||
| High team interaction × COVID-19 | 0.016*** | 0.004 | ||||||||
| (0.003) | (0.005) | |||||||||
| High customer interaction × COVID-19 | −0.015*** | −0.017*** | ||||||||
| (0.003) | (0.005) | |||||||||
| High Face2Face interaction × COVID-19 | −0.013*** | −0.006 | ||||||||
| (0.003) | (0.005) | |||||||||
| Firm-level controls | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| Firm fixed effects | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| Country-year-quarter fixed effects | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| Observations | 56,819 | 56,819 | 56,534 | 56,534 | 56,819 | 20,120 | 20,120 | 20,099 | 20,099 | 20,120 |
| Adjusted R2 | 0.918 | 0.918 | 0.918 | 0.918 | 0.918 | 0.903 | 0.903 | 0.903 | 0.903 | 0.903 |
Note(s): The dependent variable in Panel A is Capital expenditure and that in Panel B is Asset turnover. Columns (1)-(5) show the results related to developed economies and columns (6)-(10) show the results related to emerging economies. The description of all variables is presented in Table 1. The standard errors are shown in parenthesis, which are clustered at the firm level. ***, **, *denotes significance level at 1, 5 and 10% respectively
Source(s): Authors’ computations
Country-wise distribution of sample
| Country | Observations | Country | Observations |
|---|---|---|---|
| Argentina | 267 | Kuwait | 12 |
| Austria | 205 | Latvia | 34 |
| Bangladesh | 99 | Lithuania | 61 |
| Belgium | 74 | Luxembourg | 71 |
| Bermuda | 165 | Malaysia | 360 |
| Bosnia and Herzegovina | 14 | Mexico | 557 |
| Brazil | 745 | Monaco | 18 |
| Bulgaria | 112 | Netherlands | 126 |
| Canada | 702 | Nigeria | 75 |
| CavmanIslands | 47 | Norway | 376 |
| Chile | 403 | Oman | 51 |
| China | 9,394 | Pakistan | 32 |
| Colombia | 127 | Peru | 251 |
| Croatia | 287 | Philippines | 200 |
| Cyprus | 46 | Poland | 944 |
| Denmark | 262 | Portugal | 91 |
| Egypt | 360 | Republic of Serbia | 14 |
| Estonia | 89 | Romania | 95 |
| Finland | 368 | Russia | 532 |
| France | 92 | Saudi Arabia | 330 |
| Germany | 1,511 | Singapore | 297 |
| Ghana | 6 | Slovenia | 55 |
| Greece | 71 | Spain | 25 |
| HongKong | 69 | SriLanka | 310 |
| Hungary | 30 | Sweden | 649 |
| Iceland | 90 | Switzerland | 109 |
| India | 100 | Taiwan | 6,128 |
| Indonesia | 964 | Thailand | 2,467 |
| Ireland | 129 | Turkey | 894 |
| Italy | 69 | Ukraine | 8 |
| Jamaica | 31 | United Arab Emirates | 32 |
| Japan | 26,331 | United Kingdom | 165 |
| Jordan | 142 | United States of America | 18,524 |
| Kazakhstan | 55 | Vietnam | 1,852 |
| Total | 79,201 |
Source(s): Authors’ computations
Tangibility and WFH amenability
| (1) | (2) | (3) | (4) | (5) | |
|---|---|---|---|---|---|
| High remote working × COVID-19 | 0.002*** | ||||
| (0.001) | |||||
| High remote working wage × COVID-19 | 0.002*** | ||||
| (0.001) | |||||
| High team interaction × COVID-19 | 0.002*** | ||||
| (0.001) | |||||
| High customer interaction × COVID-19 | −0.004*** | ||||
| (0.001) | |||||
| High Face2Face interaction × COVID-19 | −0.004*** | ||||
| (0.001) | |||||
| Liquidity | 0.031*** | 0.031*** | 0.030*** | 0.030*** | 0.031*** |
| (0.002) | (0.002) | (0.002) | (0.002) | (0.002) | |
| Leverage | −0.001*** | −0.001*** | −0.001*** | −0.001*** | −0.001*** |
| (0.000) | (0.000) | (0.000) | (0.000) | (0.000) | |
| Profitability | 0.030*** | 0.030*** | 0.033*** | 0.033*** | 0.031*** |
| (0.007) | (0.007) | (0.007) | (0.007) | (0.007) | |
| Size | 0.014*** | 0.014*** | 0.014*** | 0.014*** | 0.014*** |
| (0.001) | (0.001) | (0.001) | (0.001) | (0.001) | |
| Constant | −0.188*** | −0.188*** | −0.186*** | −0.188*** | −0.190*** |
| (0.013) | (0.013) | (0.013) | (0.014) | (0.013) | |
| Firm fixed effects | Yes | Yes | Yes | Yes | Yes |
| Country-year-quarter fixed effects | Yes | Yes | Yes | Yes | Yes |
| Observations | 75,602 | 75,602 | 75,285 | 75,285 | 75,602 |
| Adjusted R2 | 0.195 | 0.195 | 0.194 | 0.194 | 0.195 |
Note(s): The dependent variable is Δ tangibility. COVID-19 equals 1 for Q2’2020 - Q4’2020 and 0 otherwise. The description of all variables is presented in Table 1. The standard errors are shown in parenthesis, which are clustered at the industry level. ***, **, * denotes significance level at 1, 5 and 10% respectively
Source(s): Authors’ computations
Debt and WFH amenability
| (1) | (2) | (3) | (4) | (5) | |
|---|---|---|---|---|---|
| High remote working × COVID-19 | 0.003*** | ||||
| (0.001) | |||||
| High remote working wage × COVID-19 | 0.004*** | ||||
| (0.001) | |||||
| High team interaction × COVID-19 | 0.001 | ||||
| (0.001) | |||||
| High customer interaction × COVID-19 | 0.001 | ||||
| (0.001) | |||||
| High Face2Face interaction × COVID-19 | −0.003** | ||||
| (0.001) | |||||
| Liquidity | −0.024*** | −0.023*** | −0.025*** | −0.025*** | −0.024*** |
| (0.005) | (0.005) | (0.005) | (0.005) | (0.005) | |
| Profitability | −0.049*** | −0.049*** | −0.046*** | −0.045*** | −0.048*** |
| (0.017) | (0.017) | (0.017) | (0.017) | (0.017) | |
| Size | 0.027*** | 0.027*** | 0.027*** | 0.027*** | 0.027*** |
| (0.002) | (0.002) | (0.002) | (0.002) | (0.002) | |
| Constant | −0.349*** | −0.349*** | −0.349*** | −0.349*** | −0.350*** |
| (0.023) | (0.023) | (0.023) | (0.023) | (0.023) | |
| Firm fixed effects | Yes | Yes | Yes | Yes | Yes |
| Country-year-quarter fixed effects | Yes | Yes | Yes | Yes | Yes |
| Observations | 75,877 | 75,877 | 75,559 | 75,559 | 75,877 |
| Adjusted R2 | 0.055 | 0.055 | 0.055 | 0.055 | 0.055 |
Note(s): The dependent variable is Δ debt. COVID-19 equals 1 for Q2’2020 - Q4’2020 and 0 otherwise. The description of all variables is presented in Table 1. The standard errors are shown in parenthesis, which are clustered at the industry level. ***, ** and *denotes significance level at 1, 5 and 10%, respectively
Source(s): Authors’ computations
Cash holdings and WFH amenability
| (1) | (2) | (3) | (4) | (5) | |
|---|---|---|---|---|---|
| High remote working × COVID-19 | 0.002* | ||||
| (0.001) | |||||
| High remote working wage × COVID-19 | 0.002* | ||||
| (0.001) | |||||
| High team interaction × COVID-19 | 0.001 | ||||
| (0.001) | |||||
| High customer interaction × COVID-19 | 0.003** | ||||
| (0.001) | |||||
| High Face2Face interaction × COVID-19 | 0.002 | ||||
| (0.001) | |||||
| Leverage | −0.000 | −0.000 | −0.000 | −0.000 | −0.000 |
| (0.000) | (0.000) | (0.000) | (0.000) | (0.000) | |
| Profitability | 0.129*** | 0.129*** | 0.127*** | 0.127*** | 0.129*** |
| (0.021) | (0.021) | (0.021) | (0.021) | (0.021) | |
| Size | −0.017*** | −0.017*** | −0.017*** | −0.017*** | −0.017*** |
| (0.001) | (0.001) | (0.001) | (0.001) | (0.001) | |
| Constant | 0.223*** | 0.223*** | 0.223*** | 0.224*** | 0.224*** |
| (0.009) | (0.009) | (0.009) | (0.009) | (0.009) | |
| Firm fixed effects | Yes | Yes | Yes | Yes | Yes |
| Country-year-quarter fixed effects | Yes | Yes | Yes | Yes | Yes |
| Observations | 76,871 | 76,871 | 76,552 | 76,552 | 76,871 |
| Adjusted R2 | 0.014 | 0.014 | 0.014 | 0.014 | 0.014 |
Note(s): The dependent variable is Δ cash holdings. COVID-19 equals 1 for Q2’2020 - Q4’2020 and 0 otherwise. The description of all variables is presented in Table 1. The standard errors are shown in parenthesis, which are clustered at the industry level. ***, ** and *denotes significance level at 1, 5 and 10% respectively
Source(s): Authors’ computations
Robustness test results
| Capital expenditure | Asset turnover | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| (1) | (2) | (3) | (4) | (5) | (6) | (7) | (8) | (9) | (10) | |
| High remote working × COVID-19 | 0.163*** | 0.015*** | ||||||||
| (0.029) | (0.002) | |||||||||
| High remote working wage × COVID-19 | 0.136*** | 0.014*** | ||||||||
| (0.029) | (0.002) | |||||||||
| High team interaction × COVID-19 | 0.152*** | 0.015*** | ||||||||
| (0.030) | (0.002) | |||||||||
| High customer interaction × COVID-19 | −0.118*** | −0.010*** | ||||||||
| (0.029) | (0.002) | |||||||||
| High Face2Face interaction × COVID-19 | −0.062** | −0.009*** | ||||||||
| (0.029) | (0.002) | |||||||||
| Liquidity | 0.434*** | 0.432*** | 0.420*** | 0.407*** | 0.417*** | −0.085*** | −0.085*** | −0.086*** | −0.087*** | −0.086*** |
| (0.114) | (0.114) | (0.115) | (0.115) | (0.114) | (0.009) | (0.009) | (0.009) | (0.009) | (0.009) | |
| Leverage | −0.023*** | −0.023*** | −0.024*** | −0.023*** | −0.023*** | −0.002** | −0.002** | −0.002** | −0.002** | −0.002** |
| (0.007) | (0.007) | (0.007) | (0.007) | (0.007) | (0.001) | (0.001) | (0.001) | (0.001) | (0.001) | |
| Profitability | 1.108*** | 1.123*** | 1.143*** | 1.200*** | 1.205*** | 0.129*** | 0.129*** | 0.136*** | 0.140*** | 0.136*** |
| (0.388) | (0.388) | (0.391) | (0.391) | (0.388) | (0.028) | (0.028) | (0.028) | (0.028) | (0.028) | |
| Size | 0.013 | 0.013 | 0.012 | 0.013 | 0.013 | −0.003*** | −0.003*** | −0.003*** | −0.003*** | −0.003*** |
| (0.010) | (0.010) | (0.010) | (0.010) | (0.010) | (0.000) | (0.000) | (0.000) | (0.000) | (0.000) | |
| Constant | 0.929*** | 0.933*** | 0.944*** | 0.957*** | 0.947*** | 0.314*** | 0.314*** | 0.313*** | 0.315*** | 0.317*** |
| (0.134) | (0.134) | (0.135) | (0.135) | (0.134) | (0.006) | (0.006) | (0.006) | (0.006) | (0.006) | |
| Firm fixed effects | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| Country-year-quarter fixed effects | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| Observations | 53,407 | 53,407 | 53,137 | 53,137 | 53,407 | 79,201 | 79,201 | 78,876 | 78,876 | 79,201 |
| Adjusted R2 | 0.522 | 0.522 | 0.522 | 0.521 | 0.522 | 0.915 | 0.915 | 0.916 | 0.915 | 0.915 |
Note(s): The dependent variable in columns (1)-(5) is Capital expenditure and that in columns (6)-(10) is Asset turnover. Here, COVID-19 equals 1 for Q1’2020 - Q4’2020 and 0 otherwise. The description of all variables is presented in Table 1. The standard errors are shown in parenthesis, which are clustered at the firm level. ***, ** and *denotes significance level at 1, 5 and 10% respectively
Source(s): Authors’ computations
Mean differences
| Mean | Mean difference | ||
|---|---|---|---|
| Variable | Treated | Control | p > t |
| Before PSM | |||
| Liquidity | 0.112 | 0.112 | 0.846 |
| Leverage | 0.862 | 0.682 | 0.000 |
| Profitability | 0.024 | 0.024 | 0.164 |
| Size | 13.744 | 13.595 | 0.000 |
| After PSM | |||
| Liquidity | 0.122 | 0.124 | 0.102 |
| Leverage | 0.839 | 0.815 | 0.101 |
| Profitability | 0.024 | 0.024 | 0.396 |
| Size | 13.777 | 13.841 | 0.001 |
Note(s): The top panel shows the mean of firm characteristics in the treated and control groups divided based the remote working ability before conducting the propensity score matching analysis. The bottom panel shows the mean of firm characteristics in the treated and control groups divided based the remote working ability after conducting the propensity score matching analysis. The p-value denotes the significance of the mean differences
Source(s): Authors’ computations
Propensity score matching analysis
| (1) | (2) | |
|---|---|---|
| High remote working × COVID-19 | 0.177*** | 0.013*** |
| (0.032) | (0.002) | |
| Liquidity | 0.459*** | −0.085*** |
| (0.116) | (0.009) | |
| Leverage | −0.023*** | −0.002** |
| (0.007) | (0.001) | |
| Profitability | 1.110*** | 0.130*** |
| (0.390) | (0.028) | |
| Size | 0.013 | −0.003*** |
| (0.010) | (0.000) | |
| Constant | 0.926*** | 0.315*** |
| (0.137) | (0.006) | |
| Firm-level controls | Yes | Yes |
| Firm fixed effects | Yes | Yes |
| Country-year-quarter fixed effects | Yes | Yes |
| Observations | 53,277 | 79,167 |
| Adjusted R2 | 0.522 | 0.915 |
Note(s): The dependent variable in column (1) is Capital expenditure and that in column (2) is Asset turnover. The classification of High remote working firms is based on propensity score matching analysis. The matched sample based on propensity scores is used for estimates shown in the table. The description of all variables is presented in Table 1. The standard errors are shown in parenthesis, which are clustered at the firm level. ***, ** and *denotes significance level at 1, 5 and 10% respectively
Source(s): Authors’ computations
Industry-wise clustering
| Capital expenditure | Asset turnover | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| (1) | (2) | (3) | (4) | (5) | (6) | (7) | (8) | (9) | (10) | |
| High remote working × COVID-19 | 0.176*** | 0.013* | ||||||||
| (0.055) | (0.007) | |||||||||
| High remote working wage × COVID-19 | 0.147** | 0.012 | ||||||||
| (0.053) | (0.007) | |||||||||
| High team interaction × COVID-19 | 0.144** | 0.013* | ||||||||
| (0.058) | (0.007) | |||||||||
| High customer interaction × COVID-19 | −0.124* | −0.015* | ||||||||
| (0.063) | (0.008) | |||||||||
| High Face2Face interaction × COVID-19 | −0.072 | −0.011 | ||||||||
| (0.071) | (0.007) | |||||||||
| Firm-level controls | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| Firm fixed effects | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| Country-year-quarter fixed effects | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| Observations | 53,407 | 53,407 | 53,137 | 53,137 | 53,407 | 79,201 | 79,201 | 78,876 | 78,876 | 79,201 |
| Adjusted R2 | 0.522 | 0.522 | 0.522 | 0.521 | 0.522 | 0.915 | 0.915 | 0.916 | 0.915 | 0.915 |
Note(s): The dependent variable in columns (1)-(5) is Capital expenditure and that in columns (6)-(10) is Asset turnover. COVID-19 equals 1 for Q2’2020 – Q4’2020 and 0 otherwise. The description of all variables is presented in Table 1. The standard errors are shown in parenthesis that are clustered at the industry level. ***, ** and *denotes significance level at 1, 5 and 10%, respectively
Source(s): Authors’ computations
Notes
Economic and social inequalities caused by ability/inability to access digital resources.
Table A1 provides the country-wise distribution of our sample.
The interactive fixed effects is conducted using Reghdfe package in Stata.
Figure A1 shows the sample distribution before and after propensity score matching.
References
Alarussi, A. S., & Alhaderi, S. M. (2018). Factors affecting profitability in Malaysia. Journal of Economic Studies, 45(3), 442–458. doi: 10.1108/JES-05-2017-0124.
Albuquerque, R., Koskinen, Y., & Zhang, C. (2020). Resiliency of environmental and social stocks: An analysis of the exogenous COVID-19 market crash. The Review of Corporate Finance Studies, 9(3), 593–621. doi: 10.1093/rcfs/cfaa011.
Alipour, J. -V., Fadinger, H., & Schymik, J. (2021). My home is my castle – the benefits of working from home during a pandemic crisis. Journal of Public Economics, 196, 104373. doi: 10.1016/j.jpubeco.2021.104373.
Avdiu, B., & Nayyar, G. (2020). When face-to-face interactions become an occupational hazard: Jobs in the time of COVID-19. Economics Letters, 197, 109648. doi: 10.1016/j.econlet.2020.109648.
Bai, J. (J.), Brynjolfsson, E., Jin, W., Steffen, S., & Wan, C. (2021). Digital resilience: How work-from-home feasibility affects firm performance, National Bureau of Economic Research, w28588. doi: 10.3386/w28588.
Baker, S. R., Bloom, N., Davis, S. J., Kost, K., Sammon, M., & Viratyosin, T. (2020). The unprecedented stock market reaction to COVID-19. The Review of Asset Pricing Studies, 10(4), 742–758. doi: 10.1093/rapstu/raaa008.
Barrero, J. M., Bloom, N., Davis, S. J., & Meyer, B. H. (2021). COVID-19 is a persistent reallocation shock. AEA Papers and Proceedings, 111, 287–291. doi: 10.1257/pandp.20211110.
Barry, J. W., Campello, M., Graham, J. R., & Ma, Y. (2022). Corporate flexibility in a time of crisis. Journal of Financial Economics, 144(3), 780–806. doi: 10.1016/j.jfineco.2022.03.003.
Bernanke, B., & Gertler, M. (1990). Financial fragility and economic performance. The Quarterly Journal of Economics, 105(1), 87–114. doi: 10.2307/2937820.
Bloom, N., Liang, J., Roberts, J., & Ying Jenny, Z. (2015). Does working from home work? Evidence from a Chinese experiment. The Quarterly Journal of Economics, 130(1), 165–218. doi: 10.1093/qje/qju032.
Bloom, N., Fletcher, R., & Yeh, E. (2021). The impact of COVID-19 on US firms. National Bureau of Economic Research, w28314. doi: 10.3386/w28314.
Brinca, P., Duarte, J. B., & Faria-e-Castro, M. (2020). Measuring sectoral supply and demand shocks during COVID-19. Frb St. Louis Working Paper.
Brynjolfsson, E., Horton, J., Ozimek, A., Rock, D., Sharma, G., & TuYe, H.-Y. (2020). COVID-19 and remote work: An early look at US data. National Bureau of Economic Research, w27344. doi: 10.3386/w27344.
Cherry, S. F., Jiang, E. X., Matvos, G., Piskorski, T., & Seru, A. (2021). Government and private household debt relief during COVID-19. Brookings Papers on Economic Activity, 2021(2), 141–199. doi: 10.1353/eca.2022.0002.
Chiou, L., & Tucker, C. (2020). Social distancing, internet Access and inequality. (Working Paper 26982). National Bureau of Economic Research. doi: 10.3386/w26982.
Chung, K. H., Wright, P., & Charoenwong, C. (1998). Investment opportunities and market reaction to capital expenditure decisions. Journal of Banking and Finance, 22(1), 41–60. doi: 10.1016/S0378-4266(97)00021-6.
Curran, E. (2021). Capex booms as companies prepare for a post-pandemic world. Bloomberg, Available from: https://www.bloomberg.com/news/articles/2021-09-11/global-capex-booms-as-companies-prepare-for-post-pandemic-era
Del Rio-Chanona, R. M., Mealy, P., Pichler, A., Lafond, F., & Farmer, J. D. (2020). Supply and demand shocks in the COVID-19 pandemic: An industry and occupation perspective. Oxford Review of Economic Policy, 36(1), S94–S137. doi: 10.1093/oxrep/graa033.
Dickinson, V. (2011). Cash flow patterns as a proxy for firm life cycle. The Accounting Review, 86(6), 1969–1994. doi: 10.2308/accr-10130.
Ding, W., Levine, R., Lin, C., & Xie, W. (2021). Corporate immunity to the COVID-19 pandemic. Journal of Financial Economics, 141(2), 802–830. doi: 10.1016/j.jfineco.2021.03.005.
Dingel, J. I., & Neiman, B. (2020). How many jobs can be done at home?. Journal of Public Economics, 189, 104235. doi: 10.1016/j.jpubeco.2020.104235.
Fahlenbrach, R., Rageth, K., & Stulz, R. M. (2021). How valuable is financial flexibility when revenue stops? Evidence from the COVID-19 crisis. The Review of Financial Studies, 34(11), 5474–5521. doi: 10.1093/rfs/hhaa134.
Fairlie, R., & Fossen, F. M. (2022). The early impacts of the COVID-19 pandemic on business sales. Small Business Economics, 58(4), 1853–1864. doi: 10.1007/s11187-021-00479-4.
Fazzari, S. M., Hubbard, R. G., Petersen, B. C., Blinder, A. S., & Poterba, J. M. (1988). Financing constraints and corporate investment. Brookings Papers on Economic Activity, 1988(1), 141–206. doi: 10.2307/2534426.
Fisher, I. (1933). The debt-deflation theory of great depressions. Econometrica, 1(4), 337–357. doi: 10.2307/1907327.
Gan, J. (2007). Collateral, debt capacity, and corporate investment: Evidence from a natural experiment. Journal of Financial Economics, 85(3), 709–734. doi: 10.1016/j.jfineco.2006.06.007.
Gavin, R., Harrison, L., Plotkin, C. L., Spillecke, D., & Stanley, J. (2020). The B2B digital inflection point: How sales have changed during COVID-19. McKinsey & Company, New York.
Ge, C., Huang, H., Wang, Z., Jiang, J., & Liu, C. (2023). Working from home and firm resilience to the COVID-19 pandemic. Journal of Operations Management, 69(3), 450–476. doi: 10.1002/joom.1200.
González, F. (2016). Creditor rights, bank competition, and corporate investment during the global financial crisis. Journal of Corporate Finance, 37, 249–270. doi: 10.1016/j.jcorpfin.2016.01.001.
Gopalakrishnan, B., Jacob, J., & Mohapatra, S. (2022). COVID-19 pandemic and debt financing by firms: Unravelling the channels. Economic Modelling, 114, 105929. doi: 10.1016/j.econmod.2022.105929.
Gormley, T. A., & Matsa, D. A. (2014). Common errors: How to (and not to) control for unobserved heterogeneity. The Review of Financial Studies, 27(2), 617–661. doi: 10.1093/rfs/hht047.
Guan, D., Wang, D., Hallegatte, S., Davis, S. J., Huo, J., Li, S., … Gong, P. (2020). Global supply-chain effects of COVID-19 control measures. Nature Human Behaviour, 4(6), 577–587. doi: 10.1038/s41562-020-0896-8.
Guerrieri, V., Lorenzoni, G., Straub, L., & Werning, I. (2022). Macroeconomic implications of COVID-19: Can negative supply shocks cause demand shortages?. American Economic Review, 112(5), 1437–1474. doi: 10.1257/aer.20201063.
Huang, W., & Mazouz, K. (2018). Excess cash, trading continuity, and liquidity risk. Journal of Corporate Finance, 48, 275–291. doi: 10.1016/j.jcorpfin.2017.11.005.
Huynh, N., Dao, A., & Nguyen, D. (2021). Openness, economic uncertainty, government responses, and international financial market performance during the coronavirus pandemic. Journal of Behavioral and Experimental Finance, 31, 100536. doi: 10.1016/j.jbef.2021.100536.
Kagerl, C., & Starzetz, J. (2023). Working from home for good? Lessons learned from the COVID-19 pandemic and what this means for the future of work. Journal of Business Economics, 93(1), 229–265. doi: 10.1007/s11573-022-01124-6.
Kiyotaki, N., & Moore, J. (1997). Credit cycles. Journal of Political Economy, 105(2), 211–248. doi: 10.1086/262072.
Klotz, F. (2016). Monitor, measure, incentivize: Is management as simple as that?. MIT Sloan Management Review, 57(4), Available from: https://www.proquest.com/docview/1802194730/abstract/F254F825502D4F52PQ/1
Koren, M., & Pető, R. (2020). Business disruptions from social distancing. PLoS One, 15(9), e0239113. doi: 10.1371/journal.pone.0239113.
McConnell, J. J., & Muscarella, C. J. (1985). Corporate capital expenditure decisions and the market value of the firm. Journal of Financial Economics, 14(3), 399–422. doi: 10.1016/0304-405X(85)90006-6.
Nini, G., Smith, D. C., & Sufi, A. (2009). Creditor control rights and firm investment policy. Journal of Financial Economics, 92(3), 400–420. doi: 10.1016/j.jfineco.2008.04.008.
Zhang, T., Gerlowski, D., & Acs, Z. (2022). Working from home: Small business performance and the COVID-19 pandemic. Small Business Economics, 58(2), 611–636. doi: 10.1007/s11187-021-00493-6.