We will now turn to the thirty-six suppliers (see Table 3.2) and their views on links with their customer companies. This chapter analyses the nature of these relationships by examining each of the eleven features of ACR—OCR patterns identified in chapter 1 (see Table 1.1). An overall assessment of how the ACR—OCR features fit together will be given at the end.

Transactional dependence

We start with an examination of the most tangible data, namely the number of customers each supplier trades with, and the related issue of transactional dependence. It was stated in chapter 1 that a heavy transactional dependence on a few customers is a reflection of the company's willingness to be locked into a relationship, which is a feature of OCR. By contrast, a broad customer base with minimum dependence on any one customer is a characteristic of ACR.

As Table 5.1 shows, the degree of transactional dependence is higher and the number of customers less on average in Japan than in Britain. This is consistent with interviewees' responses on the optimal situation: in Britain, 15 per cent in the PCB industry and 25 per cent in electronic assembly were often cited as the maximum allowable threshold dependence, while in Japan, having three major customers with each taking up a third of the business (‘the three pillar principle’) was offered as the ideal by many.