Globalisation has prompted an increasing number of enterprises to develop strategies for entering markets outside their home markets (Bitzenis 2004:406; Johansson 2006:5; Osland, Taylor & Zou 2001:153). According to Albaum, Duerr and Strandskov (2005:253), Halikias and Panayotopoulou (2003:340), and Leonidou et al. (2007:736), this entry into a foreign market can happen by either exporting to the market or by producing in the host country, with exporting being preferred as it requires less commitment of organisational resources and offers flexibility of managerial actions. However, enterprises face a number of challenges when attempting to initiate exports or expand their export operations (Leonidou 2004:280; Préfontaine & Bourgault 2002:123; Viviers & Calof 1999:918; Westhead, Wright & Ucbasaran 2004:501).

As regards exports from Lesotho, statistics show that, with the exception of the years 2003, 2005 and 2007, there was an increase in exports during the period 1975–2008 (Lesotho Bureau of Statistics 2010:11). Between 1975 and 1984, exports grew by 10.4%, whilst the period 1985 to 1994 saw a growth rate of 19.4%. Exports reached a high of 122.3% (Lesotho Bureau of Statistics 2010:11) in 2000 following the signing of the Africa Growth Opportunity Act (AGOA).

This growth in exports can be attributed to a number of reasons, such as South African enterprises operating in Lesotho to avoid apartheid era trade sanctions in the 1980s (Lall 2005:1004; Salm et al. 2002:14), the Lesotho National Development Corporation's (LNDC) promotion of an export-oriented manufacturing sector in the early 1990s, especially textiles (Central Bank of Lesotho 2006:1), and the signing of AGOA in 2000. AGOA is a unilateral trade preference programme that grants eligible countries in sub-Saharan Africa (SSA) duty-free access to the US market (United States Department of Commerce n.d.). In particular, the signing of AGOA has led to an inflow of foreign direct investment (FDI) from Asia into the apparel and textile industries (Central Bank of Lesotho 2006:2; MIGA 2006:23).

Clothing and garment exports have dominated Lesotho's exports (Central Bank of Lesotho 2006:2; Lesotho Bureau of Statistics 2003:3; Ng & Yeats 2004:157), contributing between 63.2% and 89.9% of total exports between 1999 and 2008. This could cause problems as the success of these products in recent years has been due to preferential access under AGOA, with AGOA set to expire in 2015 (Condon & Stern 2011:5; Congressional Research Services 2003:10; Schaefer & Markheim 2006:2). According to Sandrey et al. (2005:3), with low destination diversity the future of Lesotho's exports post-AGOA is bleak. An example of this drop in exports can be seen in 2005 when the World Trade Organisation's (WTO) Agreement on Textiles & Clothing (ATC, also known as the Multi-Fibre Arrangement (MFA)) expired. This meant that the previous buyers of garments made in Lesotho could now buy unrestricted quantities of garments from places like China, causing a drop in exports from Lesotho (Bennet 2006:165). According to the Lesotho Bureau of Statistics (2010:15, 17), Lesotho's exports are predominantly destined for either South Africa or the American continent (including the US, Canada, Chile and Mexico). For instance, in 2008 exports destined for South Africa represented 30.9% of total exports, whilst 53.5% were destined for America, leaving only 15.6% for other markets.

In view of Lesotho's low destination diversity and low diversity in export products, as well as the AGOA agreement set to expire in 2015, this study was undertaken to determine the characteristics and composition of export barriers constraining exporting by Lesotho-based manufacturing enterprises.

Over the years a number of studies have focused on export barriers (Crick & Chaudhry 2000:30; Kaleka & Katsikeas 1995:499). According to Leonidou (1995:6), the thrust of the research on export barriers occurred in the 1980s and early 1990s, even though the first empirical research appeared in the mid-1960s. Studies by Kaleka and Katsikeas (1995), Leonidou (2004), and Mahajar and Hashim (2001) focused on export constraints faced by enterprises from developed countries, with Leonidou (2004:283) in particular identifying two main groups of export barriers: internal and external. Tesfom and Lutz’s (2006:269) study conducted in developing countries supports this classification and further subdivides the two main groups into enterprise, product, industry, market and macro-environmental barriers.

Studies by Crick and Chaudhry (2000), Da Silva and Da Rocha (2001), Katsikeas and Morgan (1994), Suárez-Ortega (2003), and Sullivan and Bauerschmidt (1989) aimed to determine whether perceived export barriers were influenced by various enterprise characteristics. In particular, Crick and Chaudhry (2000:33–34) revealed a number of barriers that exhibited statistically significant differences between groups of enterprises that operate in different industries, and Sullivan and Bauerschmidt (1989:21) found that managers in the same industry had similar perceptions towards certain sets of export barriers, whilst at the same time differing in perceptions towards other barriers. As a result it was hypothesised in this study that:

Another enterprise characteristic related to perceived export barriers is enterprise size, as measured by the number of employees employed by an enterprise. Number of employees is preferred as research has revealed that managers are more willing to provide information on employment than sales volumes (Katsikeas & Morgan 1994:21). Katsikeas and Morgan’s (1994:25) findings found that smaller enterprises perceived more export problems than larger enterprises in three areas: information about the export market, product adaptation and exogenous logistical constraints. Da Silva and Da Rocha (2001:600–601) found that larger enterprises perceived themselves as being more affected by corruption in their international operations than small and medium enterprises did.

According to Katsikeas and Morgan (1994:20), and Suárez and Álamo-Vera (2005:260), larger enterprises are assumed to have greater resources, which are associated with lower levels of perceived risk in export market activities. Aaby and Slater (1989:21), however, found no relationship between enterprise size and exporting activity. It was therefore hypothesised in this study that:

Lastly, the research focused on the relationship between ownership structure and export barriers. ‘Ownership plays a key role in creating the incentives necessary to make risky decisions, including those associated with international expansion’ (Lin 2010:368). According to Lin (2010), managers may be hesitant to make the risky but necessary move since their compensation is dependent on the performance of the enterprise. Ownership structure has been operationalised in different ways. Fu, Wu and Tang (2010:13) used sources of registered capital to categorise enterprise structure. They came up with three major classes: wholly foreign-owned enterprises, joint ventures and wholly locally owned enterprises. Joint ventures also differed as to whether they had foreign control or domestic control.

Fu et al. (2010:13), in a study of Chinese manufacturing enterprises, found that wholly foreign-owned enterprises and joint ventures with foreign control have a higher probability of exporting and being persistent exporters. On the other hand, joint ventures with domestic control and wholly locally owned enterprises were not likely to export or to improve their export status.

Kim and Park (2011:5), however, differ from Fu et al. (2010:13). In their study of Korean manufacturing enterprises, they determined that concentrated ownership (irrespective of whether the dominant owners are of foreign or domestic origin) influences managers to perceive export constraints as less daunting than managers operating in enterprises that have dispersed ownership. Therefore it was hypothesised in this study that:

A judgement sampling approach was used, in that 7 out of 10 industrial areas in the country were visited and enterprises engaged in manufacturing were given a questionnaire to complete. The key-informant technique was employed, whereby key informants were managers directly responsible for international operations − in the absence of an international operations manager, the chief executive officer was approached.

Following a comprehensive literature study, the questionnaire had been developed on the basis of export barriers obtained from different authors, namely Da Silva and Da Rocha (2001), Leonidou et al. (2007), Leonidou (2004), Salm et al. (2002), Suárez-Ortega (2003), and Tesfom and Lutz (2006). The questionnaire consisted of both structured and open-ended questions (questions one and five), with the questionnaire being divided into two sections. Section A (questions one to six) comprised questions focused on enterprise characteristics, whilst Section B (questions 7.1 to 7.45) measured attitudes to export barriers using a five-point Likert scale, meaning that a total of 45 attitudes to export barriers were tested.

During the initial stage, the instrument underwent two rounds of evaluation: firstly by academics knowledgeable in the field of study, and secondly during the pilot study. A drop-and-pick-up approach was used: the researcher personally delivered the questionnaires to 171 Lesotho-based manufacturing enterprises. Of the 171, 9 enterprises failed to complete the questionnaire in the required three months, meaning that 162 usable questionnaires were received back, which translates into a 94.7% response rate.

After the questionnaires were returned, the data was first summarised using descriptive statistics. Secondly, the construct validity of the instrument was tested using an exploratory factor analysis, whilst Cronbach's alpha was used to measure the internal reliability of the instrument. Lastly, analysis of variance (ANOVA) was used to determine if enterprise characteristics influenced managers’ perception of export barriers. The significance level selected for the ANOVA was 0.05.

The aim of the study was to investigate export constraints facing Lesotho-based manufacturing enterprises in view of Lesotho's low destination diversity and low assortment of export products. In addition, the study was conducted because the AGOA agreement, which led to an inflow of FDI into apparel and textile industries, was set to expire in 2015. This study determined that Lesotho-based manufacturing enterprises face three exporting constraints: lack of financial resources for market research, lack of financial resources to finance export sales, and lack of excess manufacturing capacity for exports. The three export barriers are all internal to the enterprise and all loaded under factor 3: financial constraints. The ANOVA results further revealed that locally owned enterprises experience these constraints much more severely than do both foreign-owned and jointly owned enterprises.

With the majority of enterprises in this study being micro as well as being local, the government of Lesotho needs to pay particular attention to addressing the needs of small locally owned enterprises’ constraints in order to enable them to participate in exporting activities − especially considering that they play a pivotal role in the alleviation of poverty, job creation and economic rejuvenation. In Lesotho there are institutions such as the LNDC and the Basotho Enterprise Development Corporation (BEDCO), which are charged with export and investment promotion in the country. These agencies should ensure that local enterprises gain access to incentives that are mostly given to the large and medium enterprises, specifically those incentives that would influence their financial status, such as short-term and long-term loans, credit guarantee assistance offered to exporters, and import VAT credit facility for the local purchasing of raw material and capital goods.

Furthermore, with few jointly owned manufacturing enterprises in Lesotho, locally owned enterprises are denied the opportunity to access additional capital that foreign enterprises could bring into the venture. It also means that when foreign enterprises depart, they will not leave behind positive spillovers. Relationships between enterprises would normally form if each partner had something to offer. Whilst locally owned enterprises in Lesotho, owing to their limited capabilities and capacity, might not possess financial resources to bring into the joint venture, their local knowledge would still make them valuable to a joint-venture partner. However, the government should provide incentives to investors who opt for joint ventures so that locally owned enterprises could gain expertise and additional funds as a result of working with larger enterprises. Incentives could take the form of tax relief, low-interest loans, low prices on land leasing, as well as electricity and water incentives already widely in use in other countries.

A limitation of this study was the lack of a reliable list of manufacturing enterprises in Lesotho, as this could have affected the extent to which the findings could be generalised to all Lesotho-based manufacturing enterprises. It meant that a non-probability approach had to be adopted in selecting the sample. Also, the use of a single informant in an enterprise could possibly be seen as a problem, as it could have introduced measurement error due to individual bias.

Future research could focus on determining why micro-enterprises perceive international constraints as severely as medium and large enterprises and not in line with those of small enterprises. Also, future research should investigate why the leather industry has lower perceptions of distribution constraints than enterprises from industries classified in this study as ‘the rest’. It would also be of interest to investigate why almost 30% of enterprises that participated in the study were not interested in exporting.