On globalization being self-limiting

I went on a bit of an excursion to the oft-ignored eastern end of the island today with my dad. I have a bunch of photos, but I've done rather a lot of them recently and I have a lot of good blogging material stored up. So tomorrow it will be shots of coconuts, flowers, sun and a spider, and in the meantime it will be globalization. Those with no interest in economics, tune out now.

This whole idea arose from a conversation with my dad. A bit about my dad: he was, until he retired, the head of a large local branch of a big multinational corporation (which shall remain nameless, but you'd have heard of it). As such, although his influence was relatively small, he was privy to all the same meetings and the type of information you'd expect the CEO of a big multinational to know about. He's the kind of person whose very existence makes leftists angry; the kind of guy who says "what's wrong with a monopoly anyway?". If there are any people who can be viewed as being "on the inside" of the much-maligned globalization process, my dad is certainly among them. And he told me some interesting things about globalization from the perspective of a multinational company.

Thirty years ago, when he started in the business, the majority of multinational companies were European, and were organized along lines known as the "European model". The essence of this model was that a central company in Europe could not possibly know enough about a local market to manage factories effectively in Asia from there. Thus, they maintained a lightweight head office and concentrated on hiring very good people: they gave these people a lot of power in their local markets, and told them to get on with running the company.

This was a very successful model, and had an important side-effect. It meant that the multinational wasn't really very multinational, it was more like a federation of national companies. The big, open secret amongst global brands in the 70s and 80s was that they were global brands, not global products. Coca-cola (to pick an unrelated example) in the USA is very similar to Coca-cola in the UK, but Coca-cola in Trinidad is a very different animal, and Coca-cola in Africa is a different recipe again, despite all the fuss about the recipe being a closely-guarded secret. My dad's example was soap: the same brand of soap in Europe was a small bar with a mild scent. In the Caribbean, exactly the same brand of soap was a different size, colour and perfume. Different markets have different tastes, and so each branch of the company manufactured its own variation on the theme.

Through the 80s and 90s, however, McDonald's was in the ascendancy worldwide and with it came a new model of company: the American model, which existed before McDonalds, but which came to be known as the International model afterwards. This model, as practiced at McDonalds, was opposite to the European model: they had a very large central office where staff kept very tight control of what was going on in all the parts of the company around the world, and dictated policy and direction worldwide.

The International model was the product of two things: cheaper communications, and lower trade barriers and tariffs. As telecoms became cheaper and richer (faxes vs. telegraphs, emailed spreadsheets vs. mailed letters), it became more practical to centrally manage a widely-dispersed company. And as trade barriers went down, it became more and more economical to abandon local manufacturing in favour of regional and international manufacturing centres. Instead of 40 factories producing the product for 40 countries, 3 large factories could do the same job, and the extra shipping costs would be absorbed by the economies of scale from larger manufacturing operations. Cheaper communications also made it easier to tighten the "supply chain", a concept which multinational manufacturers like Dell obsess over constantly because so much of their money is tied up in inventory that removing the need for it, or reducing it, saves them enormous sums.

However, there was a hidden consequence of globalized manufacturing: the loss of the unacknowledged "local variations". So suddenly Coke was the same everywhere, and so was the soap and the paper and the butter and the shampoo and a thousand other products you use every day which are manufactured thousands of miles away. And as a result, global brands lose market share the more global they get. This is not speculation: this is from the horse's mouth. People in Africa don't want to drink exactly what people in Norway drink: the climate and the culture are different. Multinationals in the 90s lost market share to local brands better tuned to the needs of the local market. Globalization produced more local brands, not fewer.

Multinationals are not stupid. They are aware that this loss of variation meant a loss of market share. However, they rationalized it by noting that their savings from economies of scale and better supply chain management meant that even with lower market share, their profits were still higher, and they decided they could improve in future by spending more on marketing to convince people that what they really wanted was what everybody else wanted, not some local variation. (This is the current strategy of several multinationals, according to my dad, and it is not, so far, working very well).

Which brings me to my interesting (to me, at least) conclusion: globalization suffers a law of diminishing returns. Past a certain point, the more global and uniform your product, the less well it does because it can't take local variations into account. Therefore, globalization will not inevitably destroy local companies and local culture, simply because culture is inherently local, and so local companies will be preferred a certain amount of the time. So globalization will stabilize at a certain level and then cease its expansion.

That's not a view I've heard from any other source. Have you? And do you reckon the theory holds water?

P.S. This is my Technorati Profile, added so they'll know it's really my blog.