Impact of the COVID-19 pandemic on the South African tobacco and alcohol industries: Experiences from British American Tobacco and Distell Group Limited

Abstract

South Africa declared a State of National Disaster due to the COVID-19 pandemic, instituting a nationwide lockdown on 26 March 2020. Sale of goods and services classified as non-essential, such as tobacco and liquor, were prohibited, leading to widespread concerns about viability, job losses and investment in these industries. The study highlighted the impact of the COVID-19 lockdown on the South African alcohol and tobacco industries, taking the Johannesburg Stock Exchange-listed British American Tobacco (BTI) and Distell Group Limited (DGH) as cases. The Chow Test was utilised to determine the presence of a structural break on the BTI and DGH share prices on both the pronouncement and enactment days of the COVID-19 lockdown. Furthermore, Threshold Generalised Autoregressive Conditional Heteroskedasticity (TGARCH) (1,1) was also used to test for the effects of the COVID-19 lockdown. The sample data used was daily closing share prices from 9 May 2019 to 9 May 2020, from Google Finance. The results show a structural break on the share prices on the enactment of the 26 March 2020, COVID-19 lockdown. Furthermore, the lockdown had a negative effect on the share prices of BTI and DGH. The study concludes that the COVID-19 lockdown will have long-lasting impacts on the ability of the industries to attract financing for recovery and expansion, and existing shareholders will experience reduced earnings, if any. Policy makers should promote investment by increasing interest rates, promoting local demand and supply, and provide business support to mitigate job losses.

1. Introduction

The first case of the COVID-19 was reported on 31 December 2019, in Wuhan, China, with the World Health Organization (WHO) subsequently declaring a global pandemic on 11 March 2020 (WHO, 2020). As of 16 May 2022, the global confirmed cases were 521.357 million, with 6 288 646 deaths in over 226 countries (Worldometer, 2022). According to Currie et al. (2020), COVID-19 will likely be the worst infectious disease in this generation. However, infectious disease risks are ranked 6th in terms of impact by the World Economic Forum's Global Risk Report, way below environmental and climate change issues (Economic Forum, 2022). It therefore became a shock when the COVID-19 hit the global economy, with a global economic crisis and recession looming (Ramelli and Wagner, 2020). Authors such as Moreira and Hick (2021) as well as Fernández et al. (2022) actually equate the effects of COVID-19 to the Great Recession. Half of the world's population was put under lockdown, leading to the shutdown of financial markets, corporate offices, business and events. Business had to take drastic measures such as workforce reductions and/or closures as well as disruptions in supply chains. There was also a flight in investment and consumption amongst investors and consumers mainly from the panic of the virus spread and heightened uncertainty (Ozili, 2020; Ramelli and Wagner, 2020). Global equitable markets lost close to US$24 billion (Adenomon et al., 2020). In one week between 25 and 28 February 2020, the global stock markets lost close to US$6 trillion (Ozili, 2020). These effects negatively affected businesses and their experiences with customers, employees, citizens as well as consumer behaviours and business process functions (Accenture, 2020).

There are high uncertainties on the future economic impact of COVID-19, based on the severity of the disease and mortality rates, firm and individual behaviours as well as policy responses. Authors such as Maliszewska et al. (2020) indicate that the global Gross Domestic Product (GDP) fell by 2%, with lows of 2.5% for developing countries. Clancy et al. (2020) attest that 6.7% of the global working hours, translating to 195 million full-time employees were wiped out in the second quarter of 2020. In Southern Africa, Strauss et al. (2021) and Neva (2021) highlighted that total employment loss was 25% in more than half of the countries, with a depression which was experienced in 2020. Ashraf (2020) found stock markets responding quickly and negatively to the COVID-19 outbreak. According to Ramelli and Wagner (2020) the change in asset prices are particularly interesting in a situation of uncertainty. The changes tend to capture current expectations, and valuable opportunity to gaining insights into drivers of firm value and the working of equity markets. Asset price identifies how investors value various types of firms and their characteristics as well as how markets and financial intermediaries to process information as a disaster unfolds. Stock price reaction therefore, becomes a significant indicator (Ramelli and Wagner, 2020). According to Fernandes (2020) stock markets collapsed in March 2020, with several companies witnessing their share prices falling by more than 80%. The South African stock market fell by 40%, the 6th highest globally (Fernandes, 2020). However, this aggregated measure tends to mask the sector impact of the COVID-19 pandemic. Traditional stable sectors such as pharmaceuticals, utilities and tobacco were down by 20% or more (Fernandes, 2020). The maker of Corona beer, for instance, which is ironically called the virus beer due to the similarity in the name, lost an estimated R4.3 billion by February 2020 (Business Insider, 2020).

The first COVID-19 case in South Africa was reported on 3 March 2020 (NICD, 2020). This led to the South Africa government declaring a State of National Disaster on 15 March 2020, after 51 cases and zero deaths (GoSA, 2020a, 2020b; van Walbeek et al., 2020), and a subsequent 35 day lockdown. The interventions included the closure of business during lockdown except for essential services in what was later called Alert Stage 5 from 26 March to 16 April 2020, and then extended up to 30 April 2020 (GoSA, 2020c; van Walbeek et al., 2020). Business was not allowed to operate and increase prices. Unfair marketing tactics were prohibited as well as the sale, transportation and issuing of licences for liquor and tobacco, even in a lower Stage 4 of lock down (GoSA, 2020d, 2020e). Fernandes (2020), estimated the South Africa's economy would shrink by 3.2% due to the COVID-19 pandemic. Continuation of the lockdown up to mid-June 2020 would result in the economy shrinkage of 6.8%, with a 10.8% shrinkage when it continued up to July 2020 (Fernandes, 2020). The South African government proposed a R500 billion bailout, 80% of which was going to be sourced from various international lenders (Steyn and Klopper, 2020). As of 16 May 2022, South Africa had 3 891 793 cases, with 100 755 deaths (DoH, 2020; Worldometer, 2022).

In the South African liquor industry, Ngalonkulu (2020) highlighted that losses from the 26 March lockdown in the beer, wine, spirits and ready-to-drink industry would amount to US$908 million1 in wholesale revenue as well as US$252 million in excise tax revenue, besides negatively affecting the 100 000 ha of vineyards and US$649 million exports from wine. This sector employed between 250 000–300 000 people, who risked losing jobs due to the lock down (BUSINESSTECH, 2020). Beer controls 49% market share of alcoholic beverage in South Africa, followed by wine (16%) and spirits (15%) (Bertscher et al., 2018). Besides the 60 000 licenced sale and distributor outlets as well as the 120 000 unlicensed ones, the lockdown had a multiplier effect on other upstream and downstream business related to farming, transportation, glass and bottle manufacturing, as well as traders and small business' livelihoods, amongst others (BBC News, 2020). In terms of cigarettes, the country was losing US$3 million daily on excise duty (Curson, 2020). These losses prompted the alcohol and tobacco industry players to threaten legal action against the government's ban, setting a collision course between the state and the 31% of people aged above 15 who drink alcohol in South Africa as well as the 7 million smokers2 (BBC News, 2020; Curson, 2020). The ban also enhanced illegal liquor and tobacco trading in South Africa, with lost tax revenue from tobacco estimated at US$505 million annually (Steyn and Klopper, 2020).

Studies on the COVID-19 pandemic have mainly focussed on the epidemiology of the disease including transmission, global spread, and different interventions, with others focussing on economic consequences (Stephany and Neuhäuser, 2020). The current study focusses on the impact of the pandemic on firm performance as exhibited through the stock price. Stephany and Neuhäuser (2020) however highlighted that even though stock market information has been utilised to quantify the economic effects of infectious diseases, it comes with drawbacks. These include being prone to irrational herd behaviour with prices capturing a variety of information but being aggregated into one index. This is circumvented in the current study by isolating firm performance in the alcohol and tobacco industries in South Africa, providing a case for each of these sectors. The study significantly contributes to the literature on stock market response to disasters and crises, and the growing literature on how COVID-19 impacted stock markets, especially from a commodity perspective. For example, Ashraf (2020) examined the reaction of stock markets to the COVID-19 pandemic from 64 countries. Al-Awadhi et al. (2020) utilised firm-level share prices in China to examine the impact of the pandemic. Alfaro et al. (2020) utilised equity market values in the US and found the values decreasing.

The objective of this study was to highlight the impact of the COVID-19 pandemic on the South African alcohol and tobacco industries. These two sectors were particularly interesting as they transcend both the agricultural and manufacturing sectors. Furthermore, the South African government has viewed the two sectors as potential agents for spreading the disease. This is because alcohol consumption and smoking congregate people who also tend to share their drinks and smokes (Mungmungpuntipantip and Wiwanitkit, 2020). The two sectors were also classified as non-essential and hence were heavily affected by the 26 March COVID-19 lockdown, despite their fiscal and economic contribution to the country. This study adds to Fernandes (2020) work on the impact of COVID-19 across industries and countries by zeroing in on particular industries that have been strongly affected by disaster management policy in a developing country. Variant to Fernandes (2020) outlook of estimating the economic costs, the current study utilizes actual data to determine the real effects of the COVID-19 pandemic on the alcohol and tobacco industries. Even though a COVID-19 study on the alcohol industry was conducted in Australia (Colbert et al., 2020) this mainly concentrated on the industry marketing responses. In Germany, Koopmann et al. (2021) found that during the lockdown there was increase in alcohol and tobacco consumption especially for the middle-age. However, the study aimed at household levels. Hefler and Gartner (2021) also conducted a study and found the impossibility of totally banning the tobacco industry due to the wake of COVID-19. The current study will contribute to this existing literature on the impact of the COVID-19 pandemic on the tobacco and alcohol industries.

2. Study units: British American tobacco South Africa and Distell Group Limited

British American Tobacco South Africa (BATSA) is a secondary listing3 on the JSE, which started trading on 28 October 2008 (BAT, 2008). British American Tobacco (BAT) is a tobacco and cigarette manufacturing company founded in 1902 from a joint venture between Imperial Tobacco Company and America Tobacco Company, with headquarters in London (BAT, 2020a; 2020b). BATSA traces its roots to the United Tobacco Company in 1904, after which in 1999, due to the merger of Rothmans International and British American Tobacco, resulted in the British American Tobacco South Africa, the country's largest tobacco company (BATSA, 2020a). Some of its leading brands in South Africa are Dunhill (launched 1907), Peter Stuyvesant (1954), Kent (2006), Benson and Hedges (re-launched 2014), amongst others. Every R1 sale of BATSA products has R1.54 values added to the South African economy (BATSA, 2020b).

BATSA directly employs 10 000 people (and 72 000 indirectly from retail, transport, manufacturing, business services and agriculture), supporting 179 000 spazas, convenience stores, taverns and house-shops (BATSA, 2020c; Curson, 2020). It is involved with the purchase of 90% of tobacco produced by farmers (worth US$36 million) in the country (Curson, 2020). It has a multiplier effect, creating an additional 32 jobs per every job that BATSA creates, contributing 0.64% of the national employment and R5.84 billion in wages across the country (Quantec Research, 2016). In 2019, BATSA contributed R13 billion to the government in excise, corporate, income and other taxes, close to 27% of the total excise tax collected in South Africa (BATSA, 2020c; Curson, 2020). In 2015, BATSA had a total economic stimulus of US$2.5 billion, with US$1.4 billion in economy-wide tax revenues at 1.46% of the National Treasury's total revenue (Quantec Research, 2016). BATSA recognised the impact of COVID-19 in its proposed legal battle to fight the continued banning of tobacco products during the extended lockdown period (BATSA, 2020d). A study by van Walbeek et al. (2020) also showed that nearly half of smokers switched from major brands produced by multinational companies (MNCs) such as BATSA to local products during the 26 March 2020 lockdown, tending to negatively affect the company. These were however illegal cigarette sales because of the blanket ban on all tobacco products. BATSA accounts for 65% of South Africa's market share, nearly a quarter of the country's adult population (Dunn, 2021). According to Tobacco Tactics (2021) BATSA has remained the largest tobacco manufacturer and distributor in South Africa, with a 71.4% market share, followed by Japan Tobacco International (12%) and Phillip Morris International (8.9%). However, the lockdown resulted in loss of market share of between 10% and 14% (Cronje, 2020). Furthermore, even though BATSA had access to international markets, the global lockdown conditions of preventing cigarette sales and logistics could negatively affect the bottom line, and the share prices. Fig. 1 shows the closing stock prices for BATSA (hereafter called BTI) shares from 9 May 2019 to 9 May 2020. It is shown that on average, there was a R0.60 increase in the daily share price for the year ending 9 May 2019 (as shown from the equation in Fig. 1). The rise can be attributed to the meeting of target earnings (West, 2020). It is also evident that there was a decline in the share price from R645.17 on the 5th of March, to R498.00 on the 23rd of March increasing to a peak of R723.34 on the 28th of April 2020, and gradually decreasing from thereafter (Google Finance, 2020a).

Fig. 1.

Closing stock prices for BTI shares on the JSE

Source:Google Finance (2020a).

The South African alcohol industry is dominated by SABMiller (the largest producer globally when it merged with ABInBev), Brandhouse and Distell (Bertscher et al., 2018). Distell Group Limited is a South African company that produces and markets alcohol, with brands such as 4th Street, Viceroy, Klipdrift, Nederburg, Durbanville Hills, Hunter's Dry, Savanna, Bernini, Amarula and J C Le Roux, amongst others. Truen et al. (2011) highlight that in South Africa, SAB Miller has 85% of the market share in beer, whilst Distell accounts for 39%, 36% and 40% in wine, spirits and ciders, and ready to drinks, respectively. Distell is the third largest global producer of ciders. Distell is South Africa's leading wine producer and ranked the 10th largest exporter of wine globally in 2006, exporting 9–10% of South Africa's wine (Truen et al., 2011). Distell dominates the production of brandy, at 70% of South Africa's market share and is the fourth largest producer of brandy in the world. It also has a major share in brandy distribution due to subsidiaries such as Henry Taylor Ries, who are responsible for the wholesale of Distell brands. Distell locally sells three quarters of its brandy (Truen et al., 2011). Due to its large market share, any disruption brought about by COVID-19 would have a negative impact on Distell Group Limited (DGH's) bottom line, and ultimately share price Distell had differentiated market shares for its portfolios. For instance, Amarula had an 84% market share in Brazil (Rushton, 2015). In Taiwan, Distell's Scottish Leader was the number 2 scotch brand. In South Africa, Distell had a total market share of 13.1%, which was subdivided into 8.9% in beer, 42.9% in wine and 31.0% in spirits (Rushton, 2018). The company was formed in 2000 from the merger of Stellenbosch Farmers' Winery and Distillers Corporation and began trading on the JSE on 19 March 2001 (du Plessis and Hopkins, 2011). It employed 4936 people with an annual turnover of R26.18 billion for the year ended June 2019, an increase of 9.4% from the previous year (Distell, 2019). In 2019, Distell paid US$497 million on excise duty and US$46 million income tax. Distell first recognised the possible impact of the COVID-19 pandemic in their unaudited 6 month report ending 31 December 2019, where they identified the potential impact on their global markets (Distell, 2020a). The company halted most of its production, and only concentrated on producing alcohol for sanitisers during the COVID-19 Stages 5 and 4 lockdown periods (Gunnion, 2020). This was compounded by the suppression of their growing Chinese market as well as prohibitive tariffs (Logan, 2020). Due to the effects of the COVID-19 lockdown, the basic earnings per share (EPS) for Distell were 64.13% lower for the year ending 30 June 2020 (Businesstech, 2020a; Distel, 2020; Distell, 2021). These factors, amongst others, resulted in Distell putting two of its wine operations, Alto and Plaisir de Merle, on the market, to lighten its balance sheet and improve liquidity (Wehring, 2020). Distell (2020b) acknowledged that there had been a 2% increase in the market share of Distell products post the lockdown. However, there was a 2% decline in wine market share due to beer discounting. Fig. 2 shows the closing stock prices for DGH shares from 9 May 2019 to 9 May 2020.

Fig. 2.

Closing stock prices for DGH shares on the JSE

Source:Google Finance (2020b).

Fig. 2 shows that Distell share prices have decreased on average by R0.20 daily for the year ending 9 May 2020. Furthermore, there was a sharp decline from R134.35 on 30th January to R66.20 on the 23rd March, and fluctuating around R76.40 from thereafter (Google Finance, 2020b).

3. Methodology

3.1. Design

The study utilised the Chow Test and the Threshold Generalised Autoregressive Conditional Heteroscedasticity (TGARCH) model. The Chow Test was used to test whether the March 26, 2020, COVID-19 lockdown had an effect on the closing stock price of BTI and DGH. There was no need for a break point test as the break date was already assumed. However, another test was also conducted for the March 15, 2020, announcement of the COVID-19 lockdown.

The TGARCH (1,1) was used to test the volatility of the daily closing stock price of BTI and DGH given the COVID-19 lockdown. The daily returns from the BTI and DGH share prices, denoted by $r_t$ were used. The $r_t$ was calculated as (Tripathy and Garg, 2013):

$$r_t=\ln \left(\raisebox{1ex}{$P_t$}\!\left/ \!\raisebox{-1ex}{$P_{t-1}$}\right.\right)\ast 100$$ (1)

where $r_t$ is the return in the period $t$, $P_t$ is the daily closing share price at a particular time; $P_{t-1}$ is the closing share price for the preceding period and $ln$ is natural logarithm. Ashraf (2020) also used daily stock market prices in examining stock market reaction to COVID-19 from 64 countries. The sample period was from 9 May 2019 to 9 May 2020. The sample period was motivated by three factors: (i) it encompasses at least a full year cycle in the analysis; (ii) it includes a period when the markets were both interrupted and uninterrupted by the COVID-19 pandemic; and (iii) it includes a period within which the COVID-19 lockdown was announced and commenced, with an adequate period before and after. The daily share prices were obtained from Google Finance. The TGARCH (1,1) model was specified as follows:

$$r_t=c+\rho r_{t-1}+{\epsilon }_t$$ (2)

$$\raisebox{1ex}{${\epsilon }_t$}\!\left/ \!\raisebox{-1ex}{${\epsilon }_{t-1}$}\right.\sim N\left(0,h_t\right)$$ (3)

$$h_t={\propto }_0+{\propto }_1{\epsilon }_{t-1}^2+\delta {\epsilon }_{t-1}^2{D}_{t-1}+{\beta }_1{h}_{t-1}$$ (4)

where $h_t$ is stock return volatility, ${\propto }_0$ is a constant and $D_{t-1}$ is the dummy variable representing the COVID-19 lockdown, such that:

$$D_{t-1}=\{\begin{array}{l}1,if{\epsilon }_{t-1}^2<0,COVID19lockdownbadnews\hfill \\ 0,if{\epsilon }_{t-1}^2\ge 0,COVID19lockdowngoodnews\hfill \end{array}$$

$\delta$ (coefficient of the ARCH term) measures the leverage effect or asymmetry. A positive $\delta$ value signifies an asymmetric volatility response. ${\propto }_1$ is the GARCH coefficient term measuring the forecast variance from the last period. From [3], ${\epsilon }_t$ is a random error terms with distribution $N\left(0,{\mu }^2\right)$, where ${\mu }^2$ is the variance. It therefore, follows that from [4], $h_t\sim {\mu }^2.$ A positive COVID-19 shock, ${\epsilon }_{t-1}$, highlights a total $\delta {\epsilon }_{t-1}^2$ effect in the variance. When the COVID-19 shock is negative, the total effect in the variance is $\left({\propto }_1+\delta \right)\delta {\epsilon }_{t-1}^2$. The persistence of the COVID-19 shock depends on ${\propto }_1+\delta$ such that (Kirui et al., 2014):

$${\propto }_1+\delta \sim \{\begin{array}{l}{\propto }_1+\delta <1responsetovolatilityofreturnsreducesovertime\hfill \\ {\propto }_1+\delta =1indefinitevolatilityofreturnsispersistentovertime\hfill \\ {\propto }_1+\delta >1responsetovolatilityofreturnsincreasesovertime\hfill \end{array}$$

Random error was assumed to follow the Normal Gaussian distribution. Diagnostic tests concerning heteroskedasticity, serial correlation and normality were conducted. The TGARCH model was more appropriate as the standard GARCH models assume the negative and positive shock have same effect on volatility However, the former has larger effects (Caporale and Zakirova, 2017).

4. Results

Table 1 shows that the average share price for the study period was R575.07 for BTI and R120.97 for DGH. BTI had a maximum share price of R723.34 and a minimum of R494.59, whilst DGH had a maximum and minimum of R140.72 and R66.20, respectively. The highest share price for DGH was on the 23rd of October 2019, with a gradual reduction in share price form thereafter. A sharp decline was observed from the 3rd of February 2020, to the least share price on the 23rd of March 2020. The highest share price for BTI was observed on the 28th of April 2020.

Table 1.

Descriptive statistics.

Variable	Mean	Std. Dev.	Min	Max	Kurtosis		Skewness
					Statistic	Std. Error	Statistic	Std. Error
British American Tobacco (BTI)	575.07	57.93	494.50	723.34	−0.821	0.307	0.683	0.154
Distell Group Limited (DGH)	120.97	20.89	66.20	140.72	0.678	0.307	−1.498	0.154

Fig. 3 shows the CUSUM of squares test for the presence of a structural break in the returns from the stock prices. It is shown that during the period under study, there was a digression from the 5% significance level, indicating that there was a structural change in the model.

Fig. 3.

CUSUM of squares for BTI and DGH share prices on the 26th of March 2020.

Table 2 shows the Chow test, with 16 March, 26 March and the period between 16 and 26 March 2020, as structural breaks. In both companies, it is observed that there were no structural breaks on the 16th of March 2020 when the lockdown was announced on the night of 15 March 2020. However, structural breaks were evident at the 5% and 10% levels for BTI and DGH respectively, when the COVID-19 lockdown came into effect on the 26th of March 2020. Furthermore, it is also shown that from the period 16 to 26 March 2020, there was a structural break in the stock prices at the 1% and 5% levels for BTI and DGH share prices, respectively.

Table 2.

Chow test for the period 16 March, 26 March and 16–26 March 2020.

	BTI			DGH
	16 March 2020	26 March 2020	16–26 March 2020	16 March 2020	26 March 2020	16–26 March 2020
F statistics	2.135	3.268**	4.159***	0.132	2.712*	2.669**
Log likelihood	4.303	6.556**	16.486***	0.268	5.453*	10.703**
Wald statistic	4.270	6.536**	16.635***	0.263	5.424*	10.672**

Sig at *10%, **5%, ***1%.

Fig. 4 shows the returns plot for BTI and DGH share prices. In both plots, it is evident that there was high volatility in March 2020, when the COVID-19 lockdown was announced and came into effect. There was a 9-point reduction in the share price returns for BTI on 12th March 2020, and a 13.1-point reduction for DGH on the 16th of March 2020. The spikes in the GARCH variance graph as shown in Fig. 5 augment this. These dates correspond to a period just before and immediately after the pronouncement of the March 15, 2020, lockdown. It is also shown that the volatility continued up into April and May.

Fig. 4.

Plot of the BTI and DGH share returns.

Fig. 5.

GARCH graph with conditional variance.

Table 3 shows the results of applying the equation to give mean and variance for the returns from the closing stock price for BTI and DGH.. The mean returns for BTI were positive whilst for DGH were negative, indicating an increase and decrease in stock price, respectively. It however, shows that the mean coefficients were not significant in both cases.

Table 3.

TGARCH results of the effects of the COVID-19 lockdown on BTI and DGH share prices.

Variable	BTI		DGH
	Coefficient	Z-statistic	Coefficient	Z-statistic
Mean equation
${\propto }_0$	0.00076	0.065	−0.039	−0.466
$r_{t-1}$	0.086	1.053	−0.088	−1.193
Variance equation
${\propto }_0$	0.104**	2.193	0.148**	2.552
${\beta }_1$	0.91***	35.47	0.638***	10.939
${\propto }_1$	−0.048**	−2.074	0.204***	2.830
$\delta$	0.274***	4.755	0.383**	2.350

Model summary
${\propto }_1+\delta$	0.226($<1$)	0.587 ($<1$)
Log likelihood	−500.882	−487.400
Durban Watson statistic	2.055	1.705
Alkaike Information Criterion	4.104	3.995
Schwatz Criterion	4.190	4.080
Diagnostics
ARCH LM Test	1.100 ($p$ =0.334)	0.150 ($p$ = 0.699)
Collelogram of standardized residuals	$p$ = 0.223	$p$ = 0.524
Collelogram of standardized residuals squared	$p$ = 0.347	$p$ = 0.696
Jarque-Bera Test	23.50 ($p$ =0.000)	9.89 ($p$ =0.007)

Sig at **5%, ***1%.

Table 3 also shows that from the variance equation, the coefficient of the asymmetric term is positive (0.274 for BTI and 0.383 for DGH) and is significant at the 1% and 5% levels, respectively, indicating asymmetries in the effect of COVID-19 lockdown on share prices. It indicates that bad news has a larger effect on the volatility of the stock prices than good news because of the positive coefficient. For positive shock:

$$\widehat{h_t}=0.104+0.91{\hat{h}}_{t-1}-0.048{\hat{\epsilon }}_{t-1}^2\text{for BTI; and}$$

$$\widehat{h_t}=0.148+0.638{\hat{h}}_{t-1}+0.204{\hat{\epsilon }}_{t-1}^2\text{for DGH.}$$

For negative shock:

$$\widehat{h_t}=0.104+0.91{\hat{h}}_{t-1}+(0.274-0.048){\hat{\epsilon }}_{t-1}^2\text{for BTI}$$

$$\widehat{h_t}=0.148+0.638{\hat{h}}_{t-1}+(0.204+0.383){\hat{\epsilon }}_{t-1}^2\text{for DGH}$$

The total effect in the variance is 0.226 and 0.587 for BTI and DGH, respectively. These were less than 1, highlighting that response to the volatility of returns reduces over time. Furthermore, there was a larger variance in the volatility on share prices as induced by the COVID-19 lockdown in DGH compared to BTI. Thus, the DGH share is more volatile and riskier. The results show that the modelling of the COVID-19 lockdown was a significant determinant of the daily share prices of BTI and DGH, in that order.

The residual diagnostic tests indicate no heteroskedasticity as shown through the ARCH LM Test. There was no serial correlation as highlighted by the collelogram of standardised residuals and the errors were normally distributed as depicted by the Jarque-Bear Test.

5. Discussion

The results showed that there was volatility in the DGH and BTI share prices. This was due to anticipation by investors of the easing of restrictions and the expectation that alcohol and tobacco sales and consumption would resume after the lockdown. However this was not the case, and the lockdown was extended, likely leading to the subsequent drop in the share price caused by the continuation of alcohol and tobacco on the non-essential goods list (GoSA, 2020e). Ashraf (2020) attested to global reduction in stock prices as COVID-19 cases increased. For DGH, the decline in share price could have also resulted from the depreciation of the Rand against major currencies during this time, since they have a strong international presence. Other factors affecting the share price for BTI and DGH could have included Moody's downgrading of the country's credit rating during this time (Troskie, 2020) and the reduction in the interest rate and repo rate by the South African Reserve Bank (South African Reserve Bank, 2020). This could also have been augmented by the uncertainties surrounding the Declaration of the State of National Disaster due to the COVID-19 pandemic on the 15th of March 2020 (GoSA, 2020a).

The results further indicated that structural break effects where observed when the COVID-19 lockdown came into effect, relative to when it was announced. Speculative behaviour can explain the delayed response between the announcement and the effect of the lockdown. Ashraf (2020) also found a delayed response to stock prices due to COVID-19 confirmed cases, taking an average of 20 days to respond negatively.

There was high volatility and subsequent reduction in DGH and BTI share price returns from the day of announcement and implementation of the COVID-19 lockdown. It was also shown that the volatility continued due to speculation that the COVID-19 lockdown would continue (Curson, 2020; Troskie, 2020). Furthermore, there were situations where the wine industry players, such as DGH, would lose international markets abroad. Even though wine exporting was allowed under Stage 4 of the lockdown (GoSA, 2020e), domestic supply (the largest market for DGH products) was still suppressed. In its unaudited 3 month report ending 31 December 2019, DGH acknowledged that there was a domestic decline in wine revenue, even though from spirits, it was growing (Distell, 2020a). Furthermore, there was a 3.3% decline in Earnings Before Interest, Taxation, Depreciation and Amortisation (EBITDA), a 4.7% decline in headline earnings and a 9.0% decline in normalised headline earnings. Discounting in beer prices by competitors, increased competition, and the weakening economic conditions, resulted in a 7.8% decline in sales volume (Distell, 2020a). This could have also affected the share prices of DGH. There was a shift from MNC brands from companies such as BTI to local brands in the tobacco industry during the lockdown (van Walbeek et al., 2020). Depressed sales and fluctuation in BTI share prices was aided by a 90% increase in per stick price of cigarettes (van Walbeek et al., 2020). The status of BTI as a secondary listing could also have had an effect on its shares on the JSE. This was due to the effect of COVID-19 on the global economy, with a recession outlook, affecting liquidity and capital-raising activities, thereby having a bearing on the share price. The global market value of tobacco was estimated to be US$1 trillion by the year 2026, and by May 2020, the COVID-19 had cost US$9 billion alone in the G20 countries (Clancy et al., 2020).

There was an increase and decrease in stock prices of BTI and DGH, respectively. However, the mean coefficients were not significant in both cases. This can be a result of the stock price being affected by factors previously indicated such as Moody's downgrading of the country's credit rating, the depreciation of the Rand and the reduction in interest rate and repo rate (South African Reserve Bank, 2020; Troskie, 2020). The Moody's report that downgraded the country to full junk status referred to “continuing deterioration in fiscal strength and structurally very weak economic growth due to unreliable electricity supply, persistent weak business confidence and investment” (Businesstech, 2020b). The country's sovereign credit rating was also downgraded, raising the cost of government borrowing (Department of National Treasury, 2020). This could have caused investment phobia into the country, triggering significant capital outflows. Even though a weakened exchange rate would have resulted in exports receiving better prices, it increases value chain costs as well as costs of inputs (Troskie, 2020). Troskie (2020), predicted the South African economy would enter into a recession due to the COVID-19 lockdown, with an implication of reduced money circulation, and resultant inflation. The envisaged increase in the unemployment rate from the economic contraction could also reduce spending. However, this could be offset by the reduced interest and repo rates, improving borrowing, money supply and spending. There could also be change in consumption patterns, with a slant towards basic commodities relative to tobacco and alcohol, which are viewed as luxury goods (Coleman, 2020), even though to some extent the commodities are addictive and inelastic. Negative effects of COVID-19 could cause owners of capital and investors to shy away.

Ashraf (2020) found a negative correlation between COVID-19 cases and stock market returns. Furthermore, there was a differentiated response to the pandemic. Positive asymmetries in effect of COVID-19 lockdown on share prices were observed from the results, with reduction in volatility over time. The DGH share was more volatile and riskier than the BTI share. COVID-19 lockdown was a significant determinant of the share prices. The results were however not expected, especially for DGH given the timing of the lockdown pronouncement and enactment. This period was, ordinarily, when the wine industry was at the end of harvesting their product and storing it (Coleman, 2020). Thus, the wine portfolio of DGH was mostly affected by trade, exchange rate and international demand. The resumption of wine exports in Stage 4 of the COVID-19 lockdown should have also helped improve the share price of DGH. This however, only partially materialised. The depreciated Rand value against major currencies and the Moody's credit rating and reduced interest rates, seemed to have played a larger role in the suppressed DGH share prices after the lockdown announcement. Furthermore, there was also speculation based on the previous year of the 2019 experience of drought (Coleman, 2020; Mchunu, 2019), with investors being averse to the DGH share, thereby aiding in the fall of its price.

6. Conclusion

COVID-19 has had negative economic impact, globally. The objective of the study was to highlight the impact of the COVID-19 pandemic on the South African alcohol and tobacco industries, taking the British American Tobacco South Africa (BTI) and Distell Group Limited (DGH) as cases. The study utilised the TGARCH (1,1) model to test the effects of the pronouncement and enactment of the COVID-19 lockdown on the BTI and DGH share prices on the JSE. The sample data used was daily closing share prices from 9 May 2019, to 9 May 2020. The results revealed that the COVID-19 lockdown enactment of 26 March 2020, had a negative effect on the share prices of BTI and DGH. An initial Chow Test was conducted to ascertain whether there was a structural break on both the pronouncement and enactment days of the COVID-19 lockdown. The results showed that there was indeed a structural break on the enactment day of the policy, i.e. 26 March 2020. The study, therefore, concludes that the COVID-19 lockdown had a negative effect on the alcohol and tobacco industries, especially on investor behaviour, due to volatility in the share prices. The effects were exacerbated by economic events which were at play during the time such as the depreciation of the Rand against major currencies. The Moody's downgrading of the country's credit rating during this time and the reduction in the interest rate and repo rate by the South African Reserve Bank. It is, however, difficult to partition the effects of all these variables on the stock prices during the COVID-19 lockdown. This could be an area of further study. Furthermore, an area worth exploring is the viability of such industries in risky conditions in cognisance of the moral and ethical appendages to tobacco and alcohol consumption (Hefler and Gartner, 2020; Ioannidis and Jha, 2021). Are there trade-offs between social responsibility of reduction in alcohol and tobacco consumption at the expense of business viability. This should also be assessed in cognisance of the multiplier economic contribution of these sectors to countries such as South Africa. Not nearly enough is being done by the tobacco and alcohol industries which in many countries have some of the finest facilities from huge profits from products which are so contrary to life.

Lessons learnt are that economic and political policy should not be myopic on strictly focussing on the lockdown restricting sale production of tobacco and alcohol products and services, but should also promote investment (both domestic and foreign) through increasing the interest rates. This will tend to attract the required investment as there are higher returns to the owners of capital, but offers a negative trade-off for borrowing, especially in this financial and liquidity crisis period during a lockdown, where it becomes more expensive to borrow. Therefore, a balance will need to be struck, between economic survival from borrowing and economic expansion from investment. Other initiatives such as selling off non-productive entities as embarked upon by DGH could also help cushion tobacco and liquor industry players against the lack of liquidity during and immediately after a lockdown. Promotion of the local demand and supply can also be used to circumvent exchange rate problems during and after a lockdown to cushion business from the effect of the pandemic. Thus, there is a need to lobby the South African government to allow the sale and production of tobacco and liquor goods and services. This can be exercised under stringent lockdown conditions to allow a trade-off between continuity within the sectors as well as human and societal health. This is in view of the industries’ multiplier contributions to the economy. Tax breaks can also be utilised in promoting investment into the economy during and after a lockdown. The two companies can also diversify their portfolios. For instance, DGH can intensify alcohol production for sanitiser production which was allowed under lockdown conditions.

Funding

None.

Author contributions

Saul Ngarava was involved with conceptualization, formal analysis, methodology and writing the original draft. Abbyssinia Mushunje was responsible for supervision as well as writing review and editing. Petronella Chaminuka was responsible for supervision as well as writing review and editing. Lecadia Zhou was responsible for project administration, data curation and project administration.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.