Diversity straining

As a commodity exporter, we can spread our risk across markets, but not across prices.

New Zealand’s exports have become significantly more concentrated in recent years, and into one market in particular. China’s extraordinary growth, as well as its sheer size, and the free trade agreement that we signed in 2008, have led to about a quarter of our goods exports going to China last year (though that was down from a peak of over 30% a few years ago).

That high degree of concentration has led to frequent calls for us to diversify our export markets. Often this is couched in terms of “we should identify new markets and potential growth areas”, which is a reasonable aspiration. But sometimes it also comes with the undertone that we should reduce our exposure to China - that we shouldn’t have so many of our eggs in the basket of one country with such high economic and geopolitical risk.

We have been more concentrated than this in the past. Up until World War II, around three-quarters of our exports were going to the UK, and even as late as 1970 that share was still around one-third. Of course, given how that trade relationship worked out, there’s good reason for us to be wary of leaving ourselves in that sort of position again.

So what would the ‘right’ level of diversification look like? I don’t think there’s a clear way to answer this. But if we consider this question at a smaller scale, it can shed some light on why this is so hard to answer.

In financial economics there’s a vast body of work on portfolio theory. Here’s the basic version: imagine that you’re looking to get into the sharemarket for the first time. You’ve found two shares that have had the same average rate of return over time, and you have good reason to think that that will remain the case in the future. Also, the movements in these share prices have been somewhat correlated, but not perfectly so - sometimes they move in the same direction, other times they don’t. Which share(s) should you buy?

The answer is both - by holding a mix of the two, you’ll have the same expected rate of return over time, but with less variation around your returns than if you held either one individually. What’s more, if you can put some figures on the variance and correlation of those share prices, you can work out the mix that would give you the lowest risk for the same expected return.

The same idea applies when you add a third share to your portfolio, then a fourth, then a fifth… these shares don’t have to have the same expected rates of return as each other; you may still want to include a share with a low expected rate of return, if it has a low or negative correlation with the others. What you’ll find is that for any average rate of return you might aim for, there’s some combination of shares that has the lowest level of risk.

By the time you’ve got to a thousand different shares in your portfolio, you’ve pretty much eliminated the idiosyncratic risk; you won’t be too worried about what happens to any individual share price. What you’re left with though is systematic risk, the common thread to all share prices. You can’t diversify away this risk, no matter how many different shares you hold.

Unfortunately, there are limits to diversification when it comes to international trade. While demand (or market access) may not be strongly correlated across countries, prices are. That’s especially true if you’re exporting commodities, as we largely do - if it’s the same product regardless of where in the world it’s made, the price across countries will tend to converge on a single world price. That means the risk we face is largely systematic risk - it’s common to any market that we might operate in.

Here’s a real-life example. In 2014, Western countries placed economic sanctions in Russia after it invaded and annexed Crimea. In response, Russia placed a one-year ban (later extended to two years) on imports of agricultural products from many of those countries. New Zealand wasn’t included in that ban, though we did reduce our dairy exports to Russia in those years.

That in itself wasn’t a huge challenge for us - we were sending only a small share of our dairy exports to Russia by that time, and we were able to divert some of it to other markets. The problem was that the regions that did export a lot to Russia, particularly the European Union, also had to divert their product to other markets. And their only option was to send it to markets that we were heavily involved in - China, South-East Asia and the Middle East. World dairy prices were already quite depressed at the time, for reasons that I wrote about in my previous post, and this surge of supply into our main markets just twisted the knife further. The farmgate milk price for the 2015/16 season ended at $3.90/kg - adjusted for inflation, that’s the lowest it’s been in the last few decades.

A US-China conflict would play out in a similar way. We may be the biggest source of China’s dairy imports, but we’re not the only player in that market (indeed the US is #2). At the least, we’d have to contend with a lot more US and perhaps European milk flowing into other markets, not to mention that the economies of South-East Asia would also be smashed, and shipping could become hazardous in that part of the world. Picking a side, or even staying neutral, would not spare us from the pain of reduced global demand and a lower world price.

So yes, by all means we should look to open up access to markets where we’ve been underrepresented in the past. (The appeal of the idea of a trade agreement with India seems obvious here.) But diversification is not the answer if “great power politics” is the thing you’re really worried about.