The growth in international US dollar banking

Torrens has been busy looking for an “educational” Christmas gift for his son and was looking at the Raspberry Pi. It’s an ultra cheap, tiny computer. The makers want to encourage young teenagers to get into programming and messing about with computers rather than using them just for games. Anyway, that is a digression. While perusing the Raspberry Pi website, Torrens discovered that this British firm (it’s actually a charity) prices and sells the RPi in US dollars despite not having much of a market in the US. They explain why in the FAQs:

The components we buy are priced in dollars, and we negotiate manufacturing in dollars. Because currency markets are so volatile at the moment, we price the final board in dollars too so we don’t have to keep changing the price.

That all seems pretty sensible and it got TH thinking about the creation of money denominated in US dollars after all, if your sitting in New Zealand and want to buy one of these, you are going to have to get the dollars from somewhere, so where?

In reality, many firms that make goods for sale on world markets using imported inputs do the same and price (and transact) in US dollars rather than local currency. And, because they make that decision, it encourages others to do the same. Because RPi use US dollars so will you if you want to buy the RP for Christmas. And you will probably turn to your local bank (say the ANZ if you are in Australia or New Zealand). An earlier post talked about the amount of US dollars that the bank could create. This post talks about the driver of the global demand for US dollars (outside the US) and the implications for global banking.

The Raspberry Pi example demonstrates how international trade helps drive the demand for US dollars. It’s not a US government conspiracy. It’s just that the US economy is the world’s largest and as a big trading nation. Thus trade with the US creates a demand for its currency to make international transactions. And once that happens the network externality takes over (if one person uses a particular money it tends to encourage another person to use that money too) and before you know it, the US dollar is being used for all sorts of international transactions – using US dollars saves on having to use multiple monies unnecessarily. (To understand the network externality, just think that, for a non-US resident, once you get paid in US dollars the first thing you want to do is spend them, because they are not much use to you, but then the next person finds themselves in the same position).

Here is a chart that shows just how fast international trade has been growing relative to the size of the global economy.

Since 1980, while world GDP has grown at just under 3.5% annually, trade has been growing at about 5.7% annually. The growth in trade is interesting because it largely reflects the effects of specialisation and trade in intermediate goods (the components that the Raspberry Pi manufacturer was discussing in the above quote). Whatsmore, most of the growth in trade (and GDP) has happened outside the US (mostly in Asia) and it has created a demand for US dollars to facilitate much of it. This demand, in turn, transformed the global banking system.

Thirty years ago, global banks channeled the savings of oil rich OPEC nations via London to the US to fund the US trade deficit. But when commodity prices collapsed in the 1990’s , the banks had to find a new business model. Asia was booming (with a big blip in 1998) and trade in the region was deepening fast. East Asia was becoming an export platform to satisfy demand global demand for increasingly inexpensive manufactured goods. This involved not only shipment of not only the final good but also intra-regional trade in parts and components for everything from t-shirts to the device that you are using to read this.
Because of a large share of the global market (especially for the final goods) was the US, and because of the network externality associated with using US dollars, this trade was creating a large and rapidly growing demand for US dollars to pay for all these transactions. In the laissez faire world of international banking, the demand for this money was met, not by the US Federal Reserve, but by banks. As per the earlier post on this subject, in order to facilitate this business, the global banks increasingly sourced highly liquid US dollar assets to act as reserves for their US dollar operations. In some instances, these dollar assets were the accumulated claims on the US that the Asian economies had acquired by running persistent current account surpluses, but often the global banks borrowed them directly from US banks by issuing short term US dollar denominated debt.

Global US dollar banking became a bit like a pyramid scheme with US banks at the centre, and the world’s global banking community making up the rest. This has important consequences. You can see on the Chart what happened to global trade in 2008 when the US dollar liquidity at the heart of the pyramid scheme suddenly dried up (i.e. when the US dollar assets that banks were holding as “reserves” for their dollar operations lost their money-like quality) – global trade collapsed. If you go back to the last post – you would recognise that this sudden loss of reserves to the US dollar operations of non-US banks would have increased the perception that these banks would experience a run. And consequently, the supply of deposits that were treated as money would have been greatly reduced. Without money to pay for the different iPad (or car, or whatever) bits and pieces that were being traded and turned in to goods for global consumers, a lot of the trade just stopped. These effects were rapidly transmitted into the real economy. Growth in the Korean economy decreased by about 7 percentage points in 3% to minus 4%). If TH recalls correctly the fall in Japan’s growth rate was even higher.

Local authorities were somewhat powerless to do much about the problem. If local central banks had US dollar foreign exchange reserves, then they could have injected those into the banking system. In many instances they did, but because foreign reserves were in limited supply, people who feared a run on the global banks ran anyway knowing that those reserves would soon be eliminated. The trouble is that when there is a run, a fixed fraction of cash reserves to deposits isn’t enough. Moreover, the problems for the local banks were likely to be multiplying with the run on their US dollar business. That run was quite possibly creating significant losses for the bank as a whole, which in turn was raising concern about their local currency business. But on that side, the local central banks could stand ready to act as a lender of last resort.

Of course, local currency could have been used to finance trade, or clients could have sought to find a US bank to do business, but because of the network externality, these alternatives were likely to be more expensive. This highlights a point – using a currency other than the one used by everyone else serves to act a bit like a tax on those who use the “other” currency. It taxes production and consumption of traded goods.

What does TH take from all this. Money has real effects and national monies are not just an issue of national significance. International liquidity is a complex issue that economists still need to understand better.

Leave a Reply