Identifying the drivers of trade finance in South Africa

(1)

Identifying the drivers of trade finance in

South Africa

S.J. Boshoff

orcid.org/0000-0002-0345-0619

Dissertation

accepted in fulfilment of the requirements for the

degree Master of Commerce in Economics at the North-West

University

Supervisor:

Prof. G.W. van Vuuren

Co-Supervisor:

Prof. W. Viviers

Graduation:

Student number:

May 2020

26401207

(2)

IF I HAVE SEEN FURTHER THAN OTHERS, IT IS BY STANDING UPON

THE SHOULDERS OF GIANTS

(3)

PREFACE

This study has been assembled and completed under the book format. The work in this study was carried out as part of a master’s degree at the North-West University, and forms part of a Master of Commerce in Economics. The theoretical work undertaken in this study was done in collaboration with the School of Economic Sciences, under the supervision of Prof Gary van Vuuren and the co-supervision of Prof Wilma Viviers.

These studies represent the original work of the author and have not been submitted in any form to another university. Where use was made of the work of others, this has been duly acknowledged in the text. All data used during the course of this study were obtained from the IRESS database and any additional data obtained from third party sources were duly acknowledged within the body of the study.

The results obtained from this dissertation and the contributions they make to the existing body of knowledge are summarised in Chapter 6, which also assesses future research opportunities.

________________

SJBOSHOFF

(4)

ACKNOWLEDGEMENTS

• To my Heavenly Father, who’s mercy and love is the reason for me making it this far. I can do all things through Christ who strengthens me (Phillipians 4:13; NKJV)

• To my family, the value of your love and support cannot be measured. I would not be the man I am today, if not for the Boshoff clan.

• To my lecturers, who along the way have each contributed a part to the skills used to make this Masters possible. This Masters would not have been possible if not for all the knowledge that you’ve shown me over the years. A special thank you to Prof Derick Blaauw, who’s second year economics class started me down the path of economics, and Prof Anmar Pretorius, for never chasing me out of her office no matter how many times I bugged her with questions.

• To my friends and fellow Masters students, thank you for the memories and the support.

• To my supervisor and co-supervisor, Prof Gary Van Vuuren and Prof Wilma Viviers, thank you so much for all the help you have given me this year. This Masters would not have been possible without your amazing support.

(5)

ABSTRACT

Trade finance (or bank intermediated trade finance) plays an integral role in facilitating trade across the globe: most studies assert that trade finance (TF) forms part of more than 80% of total global trade. Although trade finance has increased in importance for policy makers after the financial crises of 2008, most studies conducted over the last decade (2009 onward) focussed on the supply side of TF and how its reduction has hampered trade. To develop a more fleshed-out and well-rounded understanding of TF, research should also be conducted into what drives the demand for TF. By applying a Robust Least Squares maximum likelihood estimation technique and using bisquares and median absolute deviation-centred (MADMED) that improves statistical dispersion of outliers around the mean, this study investigates the international and domestics variables which drive the demand for TF of several listed South African companies.

This study identified 12 instances of individually significant relationships between certain industries and the included independent variable (both domestic and international financial and economic variables). It also modelled two significant regressions for the retail industry at 1st differences and second differences. The investigation found that the

United States of America, nominal GDP, South African banks’ asset to capital ratios, and the South African Rand-British Pound exchange rate were significant at 1st differences.

The South African sovereign rating and VIX index of implied volatility variables were found to be significant at 2nd differences.

(6)

OPSOMMING

Handelsfinansiering (of bank-intermediêre handelsfinansiering) speel 'n integrale rol om handel oor die hele wêreld te vergemaklik: die meeste studies beweer dat 80% van wêreldhandel van handelsfinansiering (HF) gebruik maak. Alhoewel handelsfinansiering na die finansiële krisisse vir beleidmakers in belangrikheid toegeneem het, fokus die meeste studies wat die afgelope dekade (2009) gedoen is op die aanbodkant van HF, en hoe die vermindering daarvan die handel belemmer het. Om 'n meer gevorderde en afgeronde begrip van HF-navorsing te ontwikkel, moet daar ook ondersoek ingestel word na die vraag na HF. Deur gebruik te maak van 'n Robust Least Squares-ramingstegniek vir maksimum waarskynlikheid, asook die bisquares and median absolute afwykingsgesentreerd (MADMED), word in hierdie studie die internasionale en binnelandse veranderlikes ondersoek wat die vraag na HF van verskeie genoteerde Suid-Afrikaanse ondernemings aandryf.

Hierdie studie het 12 gevalle van individueel beduidende verhoudings tussen sekere nywerhede en die onafhanklike veranderlike (beide plaaslike en internasionale finansiële en ekonomiese veranderlikes) geïdentifiseer. Hierdie studie het ook twee beduidende regressies vir die kleinhandelbedryf gemodelleer met die eerste verskille en die tweede verskille. Die ondersoek het bevind dat die nominale BBP van die Verenigde State van Amerika, die Suid-Afrikaanse banke se bate tot kapitaalverhoudings, en die wisselkoers van die Suid-Afrikaanse Rand-Britse Pond by die eerste verskille beduidend was. Die Suid-Afrikaanse soewereine gradering en die VIX-indeks van geïmpliseerde wisselvalligheidsveranderlikes, is by die tweede verskille beduidend gevind.

Sleutelterme: Handelsfinansiering, vraag na handelsfinansiering, robuuste kleinste

(7)

CHAPTER 1 INTRODUCTION ... 1

1.1 Background and overview ... 1

1.2 Problem statement ... 1 1.3 Objectives ... 2 1.3.1 Primary objective ... 2 1.3.2 Secondary objectives ... 2 1.4 Data ... 3 1.4.1 International variables ... 3 1.4.2 Domestic variables ... 3 1.5Methodology ... 3 1.6Global trade ... 4 1.7Trade finance ... 5 1.8Chapters Breakdown ... 6

CHAPTER 2 LITERATURE REVIEW ... 8

2.1 Introduction ... 8

2.2History of trade ... 8

2.2.1 Mercantilism ... 8

2.2.2 Adam Smith’s Wealth of Nations ... 9

2.2.3 Comparative advantage ... 11

2.2.4 Paul Krugman’s New Trade Theory... 11

2.2.5 New-new trade theory ... 14

2.3 Trade finance ... 16

2.4 Problems within trade finance ... 20

2.5 Conclusion ... 22

CHAPTER 3 DATA ... 24

3.2 VIX Index of implied volatility ... 24

3.3 Financial Condition Index ... 25

3.4 Rand-Dollar Exchange rate ... 27

3.5 Nominal GDP Growth for South Africa and the USA ... 28

3.6 Sovereign credit rating for South Africa ... 29

3.7 Total exports for South Africa ... 30

3.8 Rand-Pound exchange rate ... 31

3.9 Banks’ capital ratios for South Africa and the United States... 32

3.10 A preliminary summary ... 34

(8)

CHAPTER 4 METHODOLOGY ... 39

4.1 Introduction ... 39 4.2 Data specification ... 39 4.2.1 Y variables... 39 4.2.2 X variables... 40 4.3 Multicollinearity ... 41 4.4 Stationarity ... 41

4.4.1 Levin, Lin and Chu Model ... 41

4.4.2Im, Pesaran and Shin ... 43

4.4.3 Fisher-ADF ... 44

4.4.4Chosen model and rationale ... 45

4.5 Serial correlation ... 45

4.6 Robust Least Squares ... 46

4.7 Conclusion ... 49

CHAPTER 5 RESULTS ... 50

5.2 Quality testing ... 50

5.2.1 At levels (Fisher-ADF probabilities) ... 51

5.2.2 1st differences (Fisher-ADF probabilities)... 52

5.2.3 Multicollinearity ... 53

5.3 Modelling of these data at 1st differences ... 54

5.3.1 Variables modelled at t (probabilities and coefficients) ... 55

5.3.2 Variables modelled at 𝒕 − 𝟏 (Probabilities and coefficients) ... 57

5.3.3 Final models at 1st differences ... 59

5.3.4Successful three variable regression for the retail industry grouping ... 60

5.4 Residual testing ... 60

5.4.1 Serial correlation results at 1st differences ... 61

5.5 Data modelled at 2nd differences ... 62

5.5.12nd differences (ADF probabilities) ... 62

5.5.2Variables modelled at 2nd differences (probabilities and coefficients) ... 63

5.5.3Final models at 2nd difference ... 65

5.6 Residual diagnostics... 66

5.6.1 Serial correlation results at 2nd differences ... 66

5.7 Conclusion and summary ... 66

CHAPTER 6 RECOMMENDATIONS ... 68

6.2Brief summary... 68

6.3Data problems and recommendations for future studies ... 69

6.4Policy recommendations ... 70

(9)

ANNEXURES ... 77

1.1Normality testing of variables ... 77

1.2Multicollinearity ... 81

1.3Annualised dataset ... 82

1.3.1 Stationarity testing at levels ... 82

1.3.2 Stationarity testing at 1st differences ... 86

1.3.3 Stationarity testing at 2nd differences ... 90

1.3.4 Modelling of data at 1st differences in t ... 91

1.3.5 Modelling of data at 1st differences in t-1 ... 94

1.3.6 Modelling of data at 2nd differences ... 98

1.3.7 Serial correlation ... 99

1.4 Retail industries ... 100

1.4.5 Established model for the retail sector in t ... 112

1.4.9 Additional multicollinearity test ... 122

1.5 Mining sector ... 122

1.6 Miscellaneous industries ... 144

1.7 Food producers ... 164

(10)

1.8.6 Modelling at 2nd differences ... 200

(11)

LIST OF TABLES

Table 3-1: A preliminary summary ... 34

Table 5-1: Fisher-ADF probabilities at levels ... 51

Table 5-2: Fisher-ADF probabilities at 1st differences ... 52

Table 5-3: Correlation Matrix ... 53

Table 5-4: Model probabilities at t ... 55

Table 5-5: Model coefficients at t ... 56

Table 5-6: Model probabilities at t-1 ... 58

Table 5-7: Model coefficients at t-1... 59

Table 5-8: Three variable model ... 61

Table 5-9: Serial correlations at 1st differences ... 62

Table 5-10: Fisher-ADF probabilities at 2nd differences ... 63

Table 5-11: Model probabilities at 2nd differences ... 64

Table 5-12: Model coefficients at 2nd differences ... 65

Table 5-13: Two variable model at 2nd differences... 66

(12)

LIST OF FIGURES

Figure 1-1: Global trade volume growth since 2000 (indexed in 2000 to 100) ... 5

Figure 2-1: Trade risk illustration ...17

Figure 2-2: Overview of trade finance instruments ...18

Figure 2-3: Letters of credit issued globally ...22

Figure 3-1: VIX Index of implied volatility...25

Figure 3-2: Financial condition index for South Africa ...26

Figure 3-3: USA Financial condition index...26

Figure 3-4: Rand-Dollar exchange rate ...27

Figure 3-5: Nominal GDP for South Africa...28

Figure 3-6: Nominal GDP for the USA ...28

Figure 3-7: Sovereign credit ratig for South Africa ...29

Figure 3-8: Totla exports for South Africa ...31

Figure 3-9: Rand-Pound exchange rate ...32

Figure 3-10: Banks’ capital ratios South Africa ...33

Figure 3-11 Banks’ capital ratios USA ...33

Figure 3-12: Retail industry grouping ...35

Figure 3-13: Retail industry grouping ...35

Figure 3-14: Miscellaneous producers grouping ...36

Figure 3-15: Mining and related industries grouping...36

Figure 3-16: Mining and related industries grouping...37

(13)

(14)

CHAPTER 1 INTRODUCTION

1.1 Background and overview

In 2017 a study conducted by the World Trade Organisation (WTO), it was found that its members had participated in commercial merchandise exports to the value of US$ 17.3 trillion (tn) (WTO, 2018). Currently (2019), very little is understood regarding the components and drivers of trade finance (TF) flows between countries and regions: there is no comprehensive study which connects TF volumes to robust explanatory variables (Auboin, 2016). This creates the opportunity for a deeper investigation into the drivers of TF and the nature of these variables.

The goal of this study – namely to establish a robust regression model which identifies and assesses the significance of principal explanatory constituents of South African trade finance (TF) demand – becomes even more pressing when we consider that approximately 80% of global trade relies on some version of TF (Liston & McNeil, 2013). By applying a more country-specific, firm-oriented approach, this study aims to contribute to the small pool of knowledge that has to date only identified historical banking crises as influencers and drivers of TF volumes.1 Identifying additional components of TF is of paramount importance to both policy makers (hoping to bolster trade in their respective countries), and financial service providers (who will be able to forecast the needs of their clients with greater accuracy).

1.2 Problem statement

Understanding the explanatory components of TF and how they vary across time and space is of considerable importance for estimating the temporal and geographical variations in TF, but also import-export demand/supply, and even exchange rates. Since approximately 80% of global trade relies on some version of TF (Liston & McNeil, 2013),

(15)

a more in-depth understanding of TF drivers (which are poorly understood and little-researched) will help to better explain the drivers of the bulk of South African trade.

• Knowing that TF (in addition to other factors) is an important determinant of trade flow patterns, what factors influence the volume of TF demand for individual listed firms within the South African economy?

• Knowing that large listed firms service both domestic and international customers, is a firm’s demand for TF influenced by both domestic and international variables?

1.3 Objectives

1.3.1 Primary objective

Focussing on the historical TF demands for individual listed firms, the primary objective is to use international and domestic variables in a multi-variable regression to ascertain which variables contribute to the TF demands of South African firms (i.e. which variables are demonstrably significant in explaining TF demand at certain confidence levels).

1.3.2 Secondary objectives The secondary objectives are to:

• Establish whether these TF explanatory variables are only limited to domestic variables or if international variables are also significant (i.e. is there a geographic aspect to the analysis in their effect on TF demand?).

• Establish whether economic activity in one economy influences demand for TF in another economy. In this case, will economic activity in one of South Africa’s largest trading partners (The United States of America) have an effect on the TF demand of South African firms?

• Determine whether the economic fragility of an economy can affect its demand for TF, in other words, could periods of high volatility and uncertainty also be a contributing factor to the demand for TF?

(16)

1.4 Data

The data to be used for the Y-variable will be collected from the financial statements of all the listed companies on the Johannesburg Stock Exchange that have been listed since the year 2000. These data will be comprised of the individual company’s total assets (including non-tangible assets) and trade payables, which will then be used to calculate the credit demand ratio for every year that the company has been listed. The explanatory variables (𝑋-variables) will be comprised of a host of international and domestic financial and trade-related variables:

1.4.1 International variables

• The Financial Conditions index for the USA (Financial) • VIX index of volatility (Financial)

• Dollar-Rand exchange rate (Financial) • United States total annual imports (Trade)

• United States banks’ asset to capital ratings (Financial) • Real GDP of the United States (Trade)

1.4.2 Domestic variables

• The Financial Condition Index of South Africa (Financial) • South African sovereign credit rating (Financial)

• South Africa’s total exports (Trade)

• South African banks’ asset to capital rating (Financial) • Real GDP of South Africa (Trade)

1.5 Methodology

Time series data, which contain the TF demand ratios for South African listed firms and the above-mentioned explanatory variables, will be used. The focus will be on determining

(17)

which variables have an effect on the credit demand ratios of the firms used in this study. These data will then be processed using data management software such as Excel and EViews, to test for non-stationarity (using the Fisher Augmented Dickey-Fuller test) and the existence of other time-series related data problems (such as autocorrelation). Subsequently this study will regress the data using the most appropriate method that accounts for any of the identified data problems, to ascertain whether any statistically significant relationships exist between the variables.

1.6 Global trade

In 2017 the 164 countries that form part of the World Trade Organisation exported US$ 17.43 tn in merchandise globally. With, the combined commercial exports of China, India and Singapore alone accounting for almost US$ 600 bn (WTO, 2018). These figures serve to illustrate the vast quantities of goods that are traded annually across various regions and countries and highlights the importance of this study into TF flows (Figure 1). TF contributes 80% to all trade, whether in terms of intermediated bank finance or trade insurance (Liston & McNeil, 2013). In fact, a large share of trade is activity supported by some form of TF (Auboin, 2016). This highlights the integral part that TF plays in facilitating trade across different countries.

(18)

World leaders shared this sentiment, and in 2009 at the G20 summit in London jump-started an initiative to increase their respective capacity for TF in the wake of the 2008-2009 financial crash in a bid to encourage trade and help the global economy recover (Auboin, 2016). But although the importance of TF has been highlighted in the last decade with regards to trade, no further studies have been made toward understanding what drives TF volumes. The Bank of International Settlements indicates that there currently exists no single source of TF statistics that would allow for the evaluation of TF markets sizes and composition (BIS, 2014).

1.7 Trade finance

As previously indicated, TF is estimated to form part of 80% of all trade (Liston & McNeil, 2013), but that little information is available in terms of TF composition and market size (BIS, 2014). There does, however, exist studies into historical TF volume reactions to market movements, current economic problems associated with TF access, and current market behaviour relating to TF. Relating to historical movements has been shown that TF volumes experience a large decline during periods of financial market distress, where studies showed a marked decrease in TF, especially trade related TF (Ronci, 2004). The decrease in TF stemming from the market crash of 2008-2009 has been estimated to account for between 15% and 20% of the decline in trade during that period (BIS, 2014). This phenomenon in part led to the G20 initiative that aimed to increase the capacity for TF in member states, hoping that the resulting increase in trade would assist the recovery of the global economy in the wake of the global recession (Auboin, 2016).

Another historical study of TF has also found a strong positive correlation between TF and trade volumes (Liston & McNeil, 2013). Another study echoes this claim by finding that a 1% increase in TF leads to a 0.4% increase in imports (Auboin & Engemann, 2014). A study conducted by the African Development bank (2014) found that the unmet demand for TF in Africa stood at US$ 120 bn in 2012, with a similar study by the Asian Development Bank estimating that the global unmet demand for TF could be as high as US$ 1.9 tn (Auboin, 2015).

(19)

1.8 Chapters Breakdown 1.8.1 Chapter 1

Chapter 1 sets out the research proposal as well as providing a short summary of the salient points concerning the study contents, such as the objectives, a description of the data, a discussion about the software used to analyse the data and a short summary of the relevant literature.

1.8.2 Chapter 2

Chapter 2 presents the literature review. This chapter focuses on trade history and trade finance (TF). For trade history, emphasis is placed on the evolution of trade as well as the how different theories surrounding trade have adapted and evolved to form their modern iterations. In doing so, this study establishes salient reasoning for approaching its objectives from a firm-oriented perspective. For TF, this study provides a background on the figures surrounding TF, instruments used in TF and problems pertaining to TF in its modern form.

1.8.3 Chapter 3

Chapter 3 sets out the description of the data used in this dissertation. It provides an in-depth analysis of the variables included into this study, as well as the rationale for their inclusion.

1.8.4 Chapter 4

Chapter 4 presents the methodology used. In this chapter, the models that were used to complete this study are explained and analysed. This chapter also focuses on the rationale for using certain statistical models and why certain models were not used. This includes the models used in the preliminary testing (for stationarity), the final model used for this study (robust least squares) and residual testing (serial correlation).

1.8.5 Chapter 5

Chapter 5 discusses the results obtained. In this chapter, the results derived from the modelling effort in Chapter 4 are discussed, such as the main regression results, as well

(20)

as the accuracy of the results obtained from the residual diagnostics of the main regression models.

1.8.6 Chapter 6

Chapter 6 sets out the recommendations relevant to the results obtained, providing a short summary of the whole dissertation. Included are problems that arose during various stages of the research process and provides recommendations for future research into the drivers of demand for TF.

1.9 Conclusion

To conclude, Chapter 1 provided the underlying structure of this dissertation. It is clear from the research proposal that research regarding TF is scarce, and in the case of its demand structures, essentially non-existent. Using statistical models, more information has been gathered and analysed surrounding what drives the demand for TF in South African companies. It is hoped the results can provide policy makers with the necessary information to make informed and targeted policy decisions in the future.

(21)

CHAPTER 2 LITERATURE REVIEW

2.1 Introduction

Trade has undoubtably been an integral part in the development of human civilisation over the past few millennia. In fact, by studying ancient obsidian tools unearthed in the Mediterranean, humanity can trace back the first trade efforts of man. With obsidian tools being found in villages hundreds of miles from their original source in areas that had no natural occurring obsidian deposits more than 10 000 years ago (Dixon et al, 1968). Though this information is not integral to this study, it does indicate that trade has been part of human interaction for thousands of years.

Within the last three centuries, trade and trade research have undergone a dramatic shift, starting with Smith (1778) who challenged the prevailing trade theories of the past. Since then, trade and trade theory have steadily evolved to the point where literature has recently become more focussed on the role of firms within the larger system of trade and how business is affected by trade finance. Which brings us to the topic of this study. Although firms have been highlighted as an integral part of the larger trade system, very little has been written on which variables affect an individual firm’s demand for trade finance. It is the hope of this study to address this gap in the literature by exploring the history of trade theory development, the importance of trade finance for firms, and finally what determines a firm’s demand for trade finance.

2.2 History of trade 2.2.1 Mercantilism

Mercantilism, or the mercantile system identified by Smith (1778), was the dominant trade philosophy of Western European economies from the 16th to the late 18th centuries (LaHaye, 2019). This philosophy of trade was underpinned by nationalistic sentiment aimed at increasing the wealth of one’s own nation by any means necessary, and within the camp of mercantilists two main areas of focus emerged:

(22)

i) Those who saw “bullion” (i.e. national gold reserves) as the greatest measure of wealth, and those who saw full employment as the eventual end goal of mercantilist policies (Grampp, 1952). Whether urban or agrarian, both groups used nationalistic policies to achieve these goals through a favourable balance of trade and increased domestic production. While economies saw a reversal of old feudal restrictions on trade exports that were essentially replaced by the large-scale implementation of tariffs on imports, through these policies mercantilists laboured to create an organised and protected national market and to increase their share in the global markets (Williams, 1958). In the pursuit of wealth and national growth (increasing gold reserves), mercantilists wanted to increase exports while also limiting imports into their countries. This concept of wealth was later refuted in Smith’s (1778) Wealth of Nations, which is seen as the foundation of modern economics (LaHaye, 2019).

ii) Smith (1778) makes a number of criticisms against mercantilism. First, he argued that free trade, when conducted between consenting nations, benefitted both parties. Secondly, he argued that specialisation of production lead to economies of scale and to improved manufacturing efficiency. Finally, he argued that collusion between industry and government was harmful to the general population. By 1860, England had abolished most of its mercantilist policies and had become the leading economy in Europe through the adoption of free trade policies, ushering in a new era of economic policies (Lahaye, 2019).

2.2.2 Adam Smith’s Wealth of Nations

Adam Smith, born in 1723, was a Scottish philosopher and economist who is best known for his book: An Inquiry into the Nature and Causes of the Wealth of Nations (colloquially known as The Wealth of Nations), which was first published in 1776 (Adam Smith Institute, 2019). His support for a laissez faire style of trade heralded a large-scale shift in trade policies, and its implementation saw both England and America emerge as leading economies in the late nineteenth century (LaHaye, 2019). According to Butler (2012), An Inquiry into the Nature and Wealth of Nations can be separated into five major themes:

2.2.2.1 The first theme addresses the regulation of commerce and the assertion that it is, in general, counter-productive and that the prevailing view on wealth (that gold reserves were equivalent to wealth) was incorrect. Smith (1778) believed

(23)

that the true wealth of a nation was found in the value of the goods and services that it created, and that the only way to maximise it was to not restrict the nation’s production capacity, but rather to liberalise it.

2.2.2.2 The second theme addresses production capacity, which relies on the division of labour and the accumulation of capital that makes it possible. Smith (1778) believed that efficiency could be improved by splitting production into small parts and that each part should be undertaken by specialised individuals. This specialisation would lead to surplus, which could be reinvested into production methods and capacity.

2.2.2.3 Smith’s (1778) third theme is that a country’s future is dependent on capital accumulation (i.e. the increase in assets due to profits and investment). Smith (1778) believed that the more capital is invested in improving production processes, the more wealth would be generated in the future. He advocated that countries who prosper are those that effectively grow, manage and protect their capital.

2.2.2.4 The fourth theme refers to one of Smith’s more well-known theories, namely, the ‘invisible hand theory’. Smith (1778) asserted that this system of capital accumulation and management is automatic. The ‘invisible hand’ theory stems from the idea that demand and supply within the economy would lead to the effective allocation of capital (Butler, 2012), and that the economy will always bring itself back into equilibrium. Scarce products command a higher price, and therefore a higher profit in supplying them, so producers invest capital in order to produce these products. Where there is a glut, prices and profits are low, so producers allocate their capital elsewhere. The overall idea being that industry remains focussed on a nation’s most important needs, without supervision from a national authority. This system, however, is only automatic when their free trade and markets are competitive (Butler, 2012). When governments use subsidies and tariffs to shelter industries and companies, higher prices result.

2.2.2.5 The fifth and final theme concerns competition and free markets, which are compromised from within by monopolies, tax preferences, trade controls and

(24)

Smith was a proponent for limited government to allow markets to freely resolve themselves.

2.2.3 Comparative advantage

For the past 200 years economists have expanded on the benefits of specialisation, building on the ideas set forth by Smith (1778) on how countries stand to benefit by trading with one another (Irwin, 2001). But it was Ricardo (1817) who first developed the concept of ‘comparative advantage’, the idea that two nations could still benefit from trade even if one is better at producing all goods traded between those countries (Rose, 2019). This insight deviated from theories set out by his contemporaries who focussed on absolute advantages within trade. Though one of the most important insights into the theory of international trade, the theory of ‘comparative advantage’ has seen little attention within academic literature since the 1960s (Costinot et al, 2011). The following is an excerpt taken from Ricardo (1817) in which he illustrates the concept of ‘comparative advantage’.

“To produce the wine in Portugal, might require only the labour of 80 men for one year, and to produce the cloth in the same country, might require the labour of 90 men for the same time. It would therefore be advantageous for her to export wine in exchange for cloth. This exchange might even take place, notwithstanding that the commodity imported by Portugal could be produced there with less labour than in England. Though she could make the cloth with the labour of 90 men, she would import it from a country where it required the labour of 100 men to produce it, because it would be advantageous to her rather to employ her capital in the production of wine, for which she would obtain more cloth from England, than she could produce by diverting a portion of her capital from the cultivation of vines to the manufacture of cloth”.

2.2.4 Paul Krugman’s New Trade Theory

The Nobel prize for economics was awarded to Paul Krugman in 2008 for his analysis of trade patterns and location of economic activity (Nobelprize.org, 2008). This New Trade

Theory (NTT), as it is more well-known, focussed on how countries could endogenously

develop into industrialised cores with agricultural peripheries (Krugman, 1991). Put differently, Krugman (1991) answered the questions of when and why manufacturing

(25)

became concentrated in certain regions while leaving others relatively undeveloped. Krugman’s (1991) work comprises of four sections.

2.2.4.1 Bases for regional divergence

Extensive discussions exist over the nature of the externalities that lead to the grouping of industries. Krugman (1991) indicates that most of the literature follows the original exposition of Alfred Marshall in identifying three reasons for the localisation of industries:

2.2.4.1.1 the concentration of firms in a single location creates a larger market for workers of an industry-specific skillset, ensuring lower levels of

unemployment and labour shortages.

2.2.4.1.2 centralised industries support the production of non-tradeable specialised inputs.

2.2.4.1.3 information spillover can give clustered firms better production capabilities, in comparison to their isolated counterparts.

Krugman (1991) concedes that the literature was thorough in explaining industrial localisation, but NTT focusses on answering a different question. Not why certain industries localise to a specific region, but why manufacturing, in general, congregates in a few select regions with the remaining regions playing a peripheral role.

Krugman (1991) adopted the assumption that the externalities which occasionally lead to the emergence of a core-periphery pattern, were pecuniary externalities associated with either demand or supply linkages rather than purely technological spillovers. This assumption allowed for a more concrete analysis than if the idea had allowed for external economies to arise in some invisible form (Krugman, 1991).

“To understand the nature of the postulated pecuniary externalities, imagine a country in which there are two kinds of production, agriculture and manufacturing. Agricultural production is characterised both by constant returns to scale and by intensive use of immobile land. The geographical distribution of this production will therefore be determined largely by the exogenous distribution of suitable land. Manufactures, on the other hand, may be

(26)

manufactures’ production take place? Because of economies of scale, production of each manufactured good will take place at only a limited number of sites. Other things equal, the preferred sites will be those with relatively large nearby demand, since producing near one's main market minimizes transportation costs. Other locations will then be served from these centrally located sites.” – Krugman (1991).

This investigation into demand, or more specifically, the circular nature by which demand evolves on a geographical level, explains how regions move from a primarily agricultural model to a core-peripheral model. Especially when considered with transportation costs, development and population distribution (Krugman, 1991).

2.2.4.2 A two-region model

Krugman (1991) developed a model assuming two kinds of production: agriculture (a constant returns sector tied to the land) and manufacturers (an increasing returns sector that can be located in either region). This model is a variation on the monopolistic competition framework initially proposed by Dixit and Stiglitz (1977) and is potent in its ability to yield simple institution-building treatments of seemingly intractable issues (Krugman, 1991).

2.2.4.3 Short-run and long-run equilibrium

Krugman (1991) conceded that the model lacked explicit dynamics. However, it was considered imperative that some concept of short-run equilibrium was necessary before moving onto full equilibrium. Short-run equilibrium was defined in the Marshallian way (Krugman, 1991) as an equilibrium that workers move towards the region that offers them higher real wages, leading to either convergence between regions as they move toward equality of worker/peasant rations or divergence between regions as the workers gather in one region (Krugman, 1991).

As the model moves from short-term to long-term, another important variable needs to be considered. Workers are not concerned with nominal wages but by real wages, and workers in the region of the larger population will face a lower price for manufactured goods (Krugman, 1991). In this section, Krugman (1991) further explains how these

(27)

variables interact in conjunction with one another to account for either convergence or divergence between the two modelled regions.

2.2.4.4 Necessary conditions for manufacturing concentration

This section focusses not on where equilibrium exists with regards to equal worker placement between the two regions, but on what conditions are necessary to create an equilibrium of worker concentration within one region (Krugman, 1991). By factoring in the number of firms, value of sales, wage rates, defection of workers between regions, and transportation costs, it allows for an equation that illustrates the boundary separating convergence and non-convergence. This equation simply states that in an economy characterised by high transportation costs, a small share of footloose manufacturing, or weak economies of scale, the distribution of manufacturing production will be determined by the primary stratum of non-workers. Lower transportation costs, a higher manufacturing share, or stronger economies, cause circular causation beginnings and manufacturing will concentrate in whatever region started first (Krugman, 1991).

2.2.5 New-new trade theory

One would be forgiven for thinking that the success of Krugman’s (1991) work on trade geographies would have stemmed from its empirical success. However, that was not the case, with even Krugman (1994) admitting that the state of empirical work was disappointing as it related to the New Trade Theory (Neary, 2004). This lack of empirical work changed with the introduction of general equilibrium models with economies of scale and imperfect competitions that were able to deal with the problems that were considered vital by both economic geographers and trade economists (Ottaviano, 2010).2

Ottaviano (2010) asserted that the origins of the shift in trade theory linked with the New-New Trade Theory (NNTT) could be considered Melitz’s (2003) “The impact of trade on intra-industry reallocations and aggregate industry productivity”. That refined the original concept of Krugman (1991) who researched the geography of trade by focussing on the intra-industry effect of trade on heterogenous firms. Since then researchers have focussed more on a micro-economic level, with the NNTT emphasising firms rather than sectors in trying to understand the challenges and opportunities countries face in the

(28)

modern age of globalisation (Ottaviano, 2010). This new firm-level of research has produced empirical studies by using firm-level data from various countries to create a new set of observations regarding international trade activities, with the general literature agreeing on the following stylised facts (Ciuriak et al., 2011):

2.2.5.1 It is relatively rare for firms to participate in international markets, and the export/import intensity of firms who do participate is relatively low. Coincidentally, very few firms within an industry that export use imported inputs, and exporting firms export only a small portion of their total production.

2.2.5.2 Firms that compete in the international market differ from firms that only serve the domestic market. Exporting firms that use imported inputs and engage in foreign direct investment (FDI) tend to be larger, relatively more skilled labour and capital intensive, more productive, and have been found to pay higher wages than firms not engaging in international trade.

2.2.5.3 Firms that enter the export market tend to grow faster in terms of employment and output than non-exporting firms.

2.2.5.4 Trade shows large scale dynamism in terms of changes in the size of existing trade flows and in terms of the appearance of new trade flows (new products being introduced to markets or the diversification of already exported products to new markets).

2.2.5.5 There is continuous firm entry and exit from export markets associated with the ongoing change in composition and destination of exported products.

2.2.5.6 Trade liberalisation increases firm productivity because of intra-industry allocations, instead of inter-industry reallocation. Meaning that liberalisation increases productivity by reallocation market shares and resources within an industry from low- to high-productivity firms.

2.2.5.7 The adoption of process technology is linked with a firm’s decision to export, with exporting firms usually adopting newer mass production technologies.

Identifying the evolutionary path that trade theory has taken in recent decades (1980 onwards) is thus clear. Ricardo’s (1871) theories of comparative advantages were refined

(29)

into an approach that focussed on regions and industries (Krugman, 1991), and then even further refined into modern trade theory (NNTT) which focusses on specific firms. It is this firm-level focus on trade (and trade policy) that embraces the work in this study (the study of what drives the demand for trade finance amongst firms: an important factor in encouraging and allowing firms to enter the international market).

2.3 Trade finance

Trade finance (TF) is estimated to form 80% to 90% of annual global trade (Liston & McNeil, 2013), but Kathuria and Malouche (2016) point out that most of this trade occurs without bank intermediation. This may lead to the conclusion that there is no compelling need for banks in the international trade system, but this would be incorrect. Many companies do conduct trade on an open account basis, in which exporters ship goods to customers with payment taking place after an agreed-upon period, accompanied by the use of TF instruments such as supply chain finance (SCF). With SCF being defined by the International Chamber of Commerce (ICC), as being a set of financing and risk mitigation practices to optimise the management of working capital and liquidity within the supply chain process (Global Supply Chain Finance Forum, 2016).

The ICC has seen a growth in usage of SCF by companies in recent years (ICC report, 2016), especially in instances of open account trading. What is not mentioned is that this use of open account finance is only used in instances where trade relationships already exist as well as in more developed economies (Kathuria & Malouche, 2016), as it is riskier with open account trading. This is shown in Figure 2-1, which illustrates why it is very rarely used in connection with trade to lower income countries, where the perceived political and economic risk increases the possibility of non-payment (i.e. default risk).

(30)

Figure 2-1 Trade risk illustration

Source: AXS Trade Finance 2019

This highlights the importance of banks (which provide working capital and bank-intermediated financial services to both exporters and importers) who, through intermediation and various TF instruments, help smooth over the initial stages of exporting (Kutharia & Malouche, 2016), particularly in real world circumstances where companies have no previous established business relationships nor history to draw upon (i.e by using TF instruments, such as Letter of credit, companies can alleviate the perceived risk associated with doing business in specific countries). This assistance therefore allows new exporters and importers to cultivate business relationships to the point where the contractual nature of their business dealings could improve (Defever et

al., 2016) and allow for a more versatile and flexible system of finance to develop.

Figure 2-2 provides a short summary of the various bank intermediated TF instruments shown in Figure 2-1 that can be used by both exporters and importers.

Overview of Trade Finance

Instruments

Advanced Payment

Advanced payment is a payment agreement, whereby the importer will pay the exporter prior to the shipment of any goods. This financing option is the most advantageous to the exporter as the importer carries all of the risk associated with the transaction. This financing option carries various benefits, such as the lack of complicated documentation, prompt payment and easy settlement. It should be noted that this option will most likely be used between business partners with an established business relationship.

(31)

Letters of Credit

Letters of credit (LC's) are instruments provided by third parties- usually a bank- which guarantee payment for the exporter in the scenario that the importer is unable to. Both the exporter or the importer can be the applicant for a letter of credit, whereby their bank (The issuing bank) will coordinate with the opposing parties bank (The negotiating bank) to ensure payment if a set of legal requirements are met. This financial instrument, though costly, is a good method for offloading default risk when dealing with suspect businesses (or in countries with a great deal of instability).

Transferrable letter of credit

Is a Letter of credit that is typically used when the beneficiary of the transaction are not the actual supplier of the goods being traded. They function the same as a normal Letter of credit, but have additional provisions written into the contract that allow banks to transfer funds to the designated suppliers.

Avalised Bill Also known as a Avalised bill of exchange. A bill of exchange is a document with which you as a creditor instruct your client to pay a set amount on a specified date. If accepted, the bill of exchange becomes a binding promise of payment. This bill of exchange becomes an avalised bill of exchange after it is guaranteed by the creditor's bank, thus ensuring payment.

Documentary Collection

When compared to Letters of Credit, Documentary collections are more cost effective and convenient. During a documentary collection the bank merely controls the flow of documentation and payment. Unlike with a Letter of Credit, the bank does not verify the documentation or guarantee the payment.

Open Account In an open account transaction, the exporter ships and delivers all of the traded goods before payment is made. The importer will then provide payment for the goods by an agreed upon date. This trade instrument is the opposite of an advanced payment transaction, as the exporter carries all of the risk. These transactions usually take place in very competitive markets as exporters are forced to provide favourable payment terms to beat their competition.

Figure 2-2 Overview of trade finance instruments

Source: United Nations, 2012.; KBC Bank, 2019.; Trade Finance Global, 2019.; Trade Finance Global, 2019.; Commercial Bank of Ceylon, 2020. & GBC International Bank, 2020

(32)

It is important that policy makers understand what factors contribute to driving the demand for TF in the domestic market, and what policy initiatives can be taken to help facilitate access to TF for small and medium enterprises (SMEs), especially considering that SMEs have typically been excluded from TF markets through low application approval rates (Auboin, 2015). Current literature and statistics surrounding the composition, market size and determinants of TF in particular, are scarce in some respects and non-existent in others. This dearth of information inhibits the ability of policy makers to craft effective government initiatives aimed at increasing the access to TF for these SMEs (BIS, 2014).

Little empirical work has been undertaken to explore the possible relationship between TF and trade volumes (Kohler & Saville, 2011). Studies do, however, exist which focus on historical TF flows and their reactions to market shocks. More specifically, some research investigates the supply of TF and how that supply is affected during market crashes, such as the credit crisis which triggered the 2008 to 2010 recession. TF volumes have been found to experience a large reduction in supply when faced with financial market distress, where studies found a marked decrease in TF, especially trade-related TF such as bank-intermediated transactions (Ronci, 2004). This was confirmed during the 2008-10 financial crisis where emerging markets saw a large decline in the availability in access to TF (IMF, 2009). Berman and Martin (2009) identify the transmission mechanism through which trade volumes react when a financial crisis takes place.

The first mechanism concerns the demand side of the trade volume equation in the form of an “income effect”, which captures the change in demand associated with a decrease in consumption. The second addresses the supply side in the form of a “disruption effect”, which captures the direct and indirect effect on the amount of trade-related credit available to both importers and exporters, as well as the increased risk aversion associated with economic activity. The decrease in TF stemming from the 2008-10 financial crisis has been estimated to account for 15% to 20% of the decline in trade during that period (BIS, 2014). This reduction in TF was the second largest cause of the collapse in trade volumes, which in their peak saw a fall in exports second only to that of the great depression (Contessi & Nicole, 2012). It is interesting to note, however, that TF (as a percentage of total trade) increased during this period, as TF decreased at a smaller percentage than trade values. Indeed, even facing various supply constraints, such as increased risk aversion by banks, TF volumes remained somewhat stable as businesses

(33)

were happy to off-load their default-risk to banks, even in the face of increased transaction costs (Asmundson et al., 2011).

Studies have also found total trade value and TF volumes to be positively correlated (Liston & McNeil, 2013) with a 1% increase in TF leading to a 0.4% increase in imports (Auboin &, Engemann, 2014). It is this correlation between TF access and trade volumes that forms the basis for G20 countries' plans to increase their capacity for TF in member states in the wake of the global recession, in hopes of restarting a recovery of total trade volumes that had drastically declined during that period (Auboin, 2016). The decline in trade during the 2008-10 recession was not caused by a decline in TF volumes or access, but a substantial demand shock amongst consumers (Asmundson et al., 2011).

2.4 Problems within trade finance

TF plays an integral part in driving trade, which highlights some disparities between countries and companies themselves. A study which included 45 African countries and 247 African commercial banks, found that the market for bank-intermediated TF (in 2014) was estimated to be valued at between US$ 330 bn to US$ 350 bn (African Development Bank, 2014). The worrying aspect of the study, however, was the value of unmet demand for TF on the African continent, which in 2011 was US$ 110 bn and had increased to US$ 120 bn in 2012 (African Development Bank, 2014). The Asian Development Bank has estimated that the global unmet TF demand for the preceding year could be as high as US$ 1.9 tn (Auboin, 2015).

Closer examination of TF statistics on a company level shows that large multinational companies (and especially their subsidiaries) have large advantages in the financial markets when compared to their SME competitors. With the large multinationals with a larger approval rate for their TF applications, where only 7% of applications are rejected compared to the 50% of applications that are rejected for SMEs (Auboin, 2015). This is puzzling, in that the average transaction default rate for short-term trade credit is only 0.1% of which more than half is recovered from the sale of the underlying assets (International Chamber of Commerce, 2013). This low rate of “loss given default” (LGD) can also be attributed to the fact that TF is “highly collateralised”, meaning that credit and insurance is usually provided against the sale of specific products which value can be

(34)

starting point for possible policy decisions made by lawmakers concerning the access, or lack of access, that SMEs have to TF. Especially when considering that causality has been established between a firms’ ability to export and its access to TF (Amiti & Weinstein, 2011), and that there exists a causal link between trade and TF at a macro level (Auboin & Engemann, 2014).

Other problems with TF are the current regulations set in place by the BIS, known as the Basel accords (Chauffour & Farole, 2009; Buckley et al., 2014; Kathuria & Malouche, 2016). The implementation of the Basel II accord, which replaced the original Basel accord in 2008, greatly expanded on the measurement on credit risk and introduced new techniques for the measurement for operational and market risk. It was the newly introduced measures for credit risks that were considered a problem: increasing the sensitivity for capital risk requirements in a risk fraught economic environment of the global crises placed additional pressures on banks to hold back on TF (Chauffour & Farole, 2009). This, in conjunction with the downgrades of companies during the crises, further reduced the access to TF for SMEs and banks in “risky” emerging markets. Though the regulations and risk measures set out in Basel II are still in place, there have been many additions and improvements to these measures through the introduction of Basel III. The introduction of new banking accords created its own set of problems.

“The international standards of financial regulations [Meaning Basel III] are based on the experiences of the financial crises in the US and Europe, and do not necessarily reflect the conditions of the financial sectors in Asian emerging countries” (Buckley et al., 2014).

This negative sentiment before Basel III's final implementation makes sense when it is considered that Asian markets are the destination for the majority of all letters of credits issued globally (Buckley et al., 2014), as shown in Figure 2-3

(35)

Figure 2-3 – Letters of Credit issued globally

Source: ICC report 2016

The more stringent regulations and larger capital requirements set out in the Basel III will drive up costs of TF in emerging Asian economies whose financial positions are already precarious (Buckley et al., 2014). This assertion is validated by a survey conducted by the Asian Development Bank, where 79% of banks stated that the Basel regulatory framework played a significant role in hampering TF. Banks also stated the need to reduce TF supply by 5% if BASEL III was implemented (Asian Development Bank, 2014). Another survey conducted by the International Chamber of Commerce found that 65% of respondents indicated that Basel III’s implementation was affecting the cost of funds and liquidity (Buckley et al., 2014).

2.5 Conclusion

Thus, it becomes clear that in terms of research, the knowledge into what drives the fluctuations in trade finance demand is still in its infancy. It is also clear that the lack of research as it relates to trade finance, has played an important role in hampering governments and regulatory bodies (such as the Bank of International Settlements); hampered from creating policies that decrease risk without having a negative effect on individual firm’s access to trade finance. The lack of research has also been a factor, hampering firms’ ability to trade, which has also been shown to negatively affect economic

(36)

understanding of what affects the demand for trade finance by firms, and allow them to better wield policy in facilitating trade through trade finance.

(37)

CHAPTER 3 DATA

3.1 Introduction

The lack of literature on the subject of TF gives rise to another problem as it relates to the modelling and regressing of these data. More specifically, there is not an established set of explanatory variables that can be regressed against data to establish what may drive the demand for TF exists to subsequently help in determining what factors drive trade finance demand. Because of the data limitations, it was decided that a more focussed study would be the most appropriate in which data from South Africa and one of its largest trading partners would be used. China was excluded because of concerns surrounding the accuracy of its published national data (Shao & Standing, 2014) and Great Britain because of possible multicollinearity of financial data. Instead data from the USA were used as most of these are easily available and some indices are calculated locally (in the USA). This study will draw heavily from previously established variables that were tested in the working paper of Garralda et al. (2015-2018), that included the following explanatory variables, both international (to account for the effect of global market conditions) and domestic (to identify regional specific determinants of TF):

3.2 VIX Index of implied volatility

The original idea for creating a set of instruments that represent volatility (i.e. an Index) was first developed by Brenner and Galai (1986). The index arose from the necessity of having an efficient tool for hedging against changes in volatility (Dondoni et al., 2018). Building on this concept, the Chicago Board Options Exchange (CBOE) launched the CBOE Market Volatility Index or VIX in 1993, which represented the volatility of the Standard & Poors (S&P) 100 equity market. In 2003, the VIX calculations were revised and the underlying index was changed to the S&P 500 Index (Dondoni et al., 2018). Currently (2019), the VIX index is based on a calculation that is designed to produce a measure of constant 30 day expected volatility of the US stock market, derived from real-time, mid-quote prices of the S&P 500 Index call and put options (CBOE, 2019).

(38)

Figure 3-1 – VIX Index of implied volatility Source: CBOE 2019

Figure 3-1 shows that markets experienced high levels of volatility in the period that spanned the financial crises and the following recession that started in late 2008. This spike in volatility should match a similar decrease in TF, as the literature has asserted the inverse relationship between market uncertainty and the access to TF(Ronci, 2004; IMF, 2009).

The VIX Index is one of the most widely used measures of volatility (CBOE, 2019) and its connection to stock market risk and volatility should create a negative correlation with TF volumes and subsequently have a detrimental effect on the demand for TF in the market.

3.3 Financial Condition Index

The Financial Condition Index (FCI) is a comprehensive index construct based on several variables, such as exchange rates and asset prices (Zheng & YU, 2014). In the past, various methods have been used to create FCIs, with the most popular approaches being the weighted-sum approach and the principal component approach (Hatzius et al., 2010). In the first approach, the weights of each financial indicator are assigned according to their estimated impact on real GDP growth in a vector autoregressive (VAR) or structural macro-economic model (Gumata et al., 2012). In the second approach, the FCI reflects a common factor, which is extracted from a group of financial indicators, and captures the greatest common variation amongst them (Gumata et al., 2012). The FCI is included as an explanatory variable because of its use as a forecasting tool for economic trends and its use by central banks in shaping domestic monetary policy (Zheng & Yu, 2014). The

18.07 0 10 20 30 40 50 60 70

Jan-00 Jan-02 Jan-04 Jan-06 Jan-08 Jan-10 Jan-12 Jan-14 Jan-16 Jan-18 Jan-20

C BO E V IX in d ex

(39)

FCIs for South Africa and the USA embrace secondary data, gathered from a third-party data provider and are shown in Figure 3-2 and 3-3. It is expected for both indices to have an inverse relationship with TF demand, as more risk would incentivise companies to offload some of their risk to their banks.

Figure 3-2 – Financial Condition Index for South Africa Source: Quantec 2019

Figure 3-3 – USA Financial Condition Index Source: Federal Reserve Bank of St Louis 2019

-20 -16 -12 -8 -4 0 4 8 12

SA Fi n an cia l C on d it ion In d ex -1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0

U SA fin an cia l con d it io n in d ex

(40)

3.4 Rand-Dollar Exchange rate

The Dollar (value of the Dollar is expressed in South African Rand) is included as an explanatory variable because of its direct international link as the most traded currency, with a daily average volume of US$ 2.2 tn (Hall-Smith, 2018). In addition, most commodities are priced in US dollars (IG, 2019) and since South Africa is heavily reliant on a large amount of commodity exports (ITC Trademap, 2019), this should allow our study to capture the effect that currency fluctuations may have on the domestic demand for TF in South Africa.

Figure 3-4 – Rand Dollar exchange rate Source: Exchangerates.org.uk 2019

The increased political uncertainty (from recent events such as the State capture inquiry) that has plagued the South African economy, should be the largest contributor in the Rand’s slow decline over the last decade. This increase in political uncertainty that has been quantified by the World Economic Forum and their Global Competitiveness Reports spanning the period 2010-2014, shows a marked decline in institutional power that corroborates the decline seen in Figure 3-4 (WEF, 2010; WEF, 2011; WEF, 2013)3. These are secondary data gathered from third-party data providers. It is expected for the Rand-Dollar exchange rate to have a positive relationship with TF demand, as the increased risk and weakened Rand to serve as incentive for companies to seek out TF to engage in trade and decrease their risk.

WEF – World Economic Forum 0.06 0.08 0.10 0.12 0.14 0.16 0.18

C o st o f ZAR1 in U S$

(41)

3.5 Nominal GDP Growth for South Africa and the USA

Nominal GDP growth (Bruno & Shin, 2014) is included in the regression as an explanatory variable, because merchandise trade tends to fluctuate proportionally in relation to changes in national income (Asmundson et al., 2011).

Figure 3-5 – Nominal GDP for South Africa Source: CEIC 2019

Figure 3-6 – Nominal GDP for the USA Source: CEIC 2019

Figures 3-5 and 3-6 show that the South African and United States economies have had

2 3 4 5 6 7 8 9 10 11

Jan-00 Jan-02 Jan-04 Jan-06 Jan-08 Jan-10 Jan-12 Jan-14 Jan-16 Jan-18

$ H u n d re d B ill io

n

_{South Africa nominal GDP}

10 12 14 16 18 20 22

Jan-00 Jan-02 Jan-04 Jan-06 Jan-08 Jan-10 Jan-12 Jan-14 Jan-16 Jan-18

$ T ri lli o n

USA nominal GDP

(42)

the only instance of a large contraction within the market. It is suspected that the performance of both these economies would have a large influence on how companies develop their trade strategies, with this paper believing that there would exist a positive relationship between economic growth for both these economies as well as the demand for TF for South African companies. Because of the highly trending nature of these data, it would be prudent to confirm the stationarity of these data to avoid the potential regression bias that becomes a problem in instances of non-stationarity. These data are secondary and was collected from a third-party source.

3.6 Sovereign credit rating for South Africa

Sovereign credit rating is included into this study as an explanatory variable following recent literature on the links between sovereign credit ratings and the banking sector (Garralda et al., 2015-08), as well as the study published by Arteta & Haley (2008) that found statistically significant and robust effects on the non-financial sector originating from sovereign credit problems.

Figure 3-7 – Sovereign credit rating for South Africa Source: Fitch (2019)

Since the inception of democracy in South Africa in 1994, the country had enjoyed successive upgrades in its credit rating from below to well into investment grade (SARB, 2017). This trend reversed in 2012 when all three credit ratings agencies (i.e S&P, Fitch and Moody’s) downgraded South Africa’s long-term credit outlook (SARB, 2017), as seen in Figure 3-7. This trend has continued with subsequent downgrades by every ratings agency, with both S&P and Fitch having downgraded South African bonds into the

15 16 17 18 19 20 21

C red it R at in g as n u m e ri ca l v al u e

(43)

speculative grade (April 2017). Moody’s is currently the only ratings agency that has kept South African bonds within the investment grade (SARB, 2017). These data have been adapted from their original rating scale by ascribing a numerical value that corresponds to a rating within the original scale (Annexure 1.8.9 – p203). This allows this study to regress South Africa’s change in sovereign credit ratings as far back as January 2000. These are secondary data gathered from third party providers, and we expect a direct relationship between sovereign credit ratings and the demand for TF to exist. This is hypothesised because of the direct effect that credit downgrades have on the financial sector and the real economy at large through the credit rating channel (Almeida et al., 2017).

3.7 Total exports for South Africa

It has been well established in the literature that trade finance (TF) supply and trade volumes are positively correlated (Liston & McNeil, 2013), with trade volumes seeing large declines during periods of credit distress (Ronci, 2004; IMF, 2009; Contessi & Nicole, 2012; Auboin & Engemann, 2014). It has also been found that past experiences with TF have positive effects in the present (Dary & Haruna, 2019). What has not been established is whether past experiences by companies, as relating to general trade itself, could affect the demand for TF by companies. Stated differently, can past export performance affect the general demand for TF amongst domestic companies? To that end, this study will use a lagged variable (by one period) of total exports for South Africa, to simulate past experiences with trade within the regression analysis.

Identifying the drivers of trade finance in South Africa