On the face of it, it makes no sense that the international flower industry should be headquartered in the Netherlands. The feeble sunshine and predisposition for a large number of rainy days would not make the Netherlands the first choice for anyone starting a flower-growing business today – if not for the fact that the business, and its integral supply chains, are already there. This is a huge competitive advantage for a new entrant, who can benefit from such things as the sophisticated Dutch flower auctions, the flower-growers’ associations and advanced research centres.

Academic Michael Porter uses this very example to illustrate his cluster theory of trade development, whereby whole supply chains “cluster” together. Another well-known cluster is the auto manufacturing industry in Michigan in the US. Over 50 per cent of North American auto companies are based in Michigan, and 46 of the top 50 global auto suppliers have operations in the state. Further south in the US, around Dalton in Georgia, over 90 per cent of all functional carpets are produced. It is why Dalton is called the “carpet capital of the world”.

It’s partly because of the cluster effect that we believe Chinese exports will be sufficiently competitive to ride out the economic slowdown and emerging market uncertainty. Clustering is certainly nothing new to China – regional specialisations have always existed. A few centuries ago, tailors from Cixi (in the Zhejiang sub-provincial city of Ningbo) were known for their skills in clothing manufacturing and even now the area is a major centre for textile manufacturing.