Wednesday, August 17, 2016

RBA wants one-sided coin on foreign capital

The below excerpt is from an interview with RBA Governor Glenn Stevens on 15th August 2016, on the topic of foreign capital. It made the front page of the nation’s most popular newspaper. Read it closely, particularly the bold part.
That’s not something that the Reserve Bank can wave a wand and make go away. Australia wants to be open to foreign capital. That’s our national philosophy. I think in that discussion, it would be helpful to think about the kind of foreign capital we want. 

Foreign capital that builds new assets, like some of the capital that funded the mining boom. That’s one thing. Foreign capital that buys up the existing assets, I’m not saying that we should be closed to that, but that’s not creating new capital for the country, that’s just altering the allocation of who owns the capital that’s here now. 

And I think when we all talk about – you know, we want capital inflow, we can probably have a bit of nuance and subtlety over what kind of inflow we mean and ask ourselves whether we’re attractive enough to the kind of capital that actually builds new assets.
I think the Governor is confused here. He appears to be reiterating some all-too-common economic nonsense by using the word capital with two different meanings in the same breath. By doing so he seems to want an outcome that is beyond the realm of accounting reality.

He says that capital, in its strict economic meaning of machines and equipment, is good to have foreigners invest in. These help create new productive “assets” [1]. Then he says that capital, with its financial meaning of non-current assets (like bonds, equities, and property), is not good to have foreigners invest in, because it is just a transfer of ownership of existing assets.

This is weird, for two reasons.

First, economic capital is just a type of good. Foreign economic capital is therefore just the importation of machines and equipment from abroad. To be clear, in this analysis I will use the term Good Foreign Capital to mean imports of machines and equipment. But mining booms aren’t funded by gifts of machinery. They are built with them, but these machines need to be paid for.

Second, if you import more Good Foreign Capital than you export, the gap must necessarily be made up by sales of assets to foreigners, or “altering the allocation of who owns the capital that’s here now”, in the words of the Governor. You can begin to see the problem. I will call selling existing assets to foreigners Bad Foreign Capital.

In the above table, showing Australia’s 2010/11 external accounts, Good Foreign Capital is an import, with a negative sign. Because all accounts balance, this import of mining machinery can either be balanced by exports, or Bad Foreign Capital (labelled Direct and Portfolio investment abroad), both of which have positive signs [2].

If Good Foreign Capital is balanced by exports of other goods and services, we are in a world with a zero foreign investment on balance. Inwards Bad Capital equal outwards Bad Capital. Overall, there is no Bad Foreign Capital. But there is also no net Good Foreign Capital either. If Glenn Stevens wants to balance trade, he should just say it.

However, if Good Foreign Capital is paid for with Bad Foreign Capital, we are in a world of with a positive capital account balance (and a negative trade balance), as Australia has been for all but a handful of the last 150 years. Economists have mostly seen this as a good thing, by justifying the Bad by its offsetting Good. You cannot have a trade deficit, or net Good Foreign Capital, without paying for it with Bad Foreign Capital.

A simple example could help clarify.

Imagine a local miner who has the rights to extract coal in an area, but not the local funding to build  up the mine with the necessary equipment. They enter into a joint venture with a foreign company. That company supplies the foreign-made mining equipment for a share of the equity in the project.

This could be the type of example Glenn Stevens has in mind when he talks about Good Foreign Capital. But in fact, it is an example of using Bad Foreign Capital funding Good Foreign Capital. The imported machines are funded by the sale of the equity stake by the local miner, which is totally non-productive and a mere “allocation of who owns the capital that’s here now”. The two types of capital are two sides of the same coin in this example. And they are also two sides of the same coin at a macro level, given the entrenched nature of Australia’s foreign position as a net seller of assets which funds its trade deficit.

What Glenn Stevens seems to be saying is that he wants to increase exports to pay for Good Foreign Capital, which would bring the capital and current accounts closer to each balancing on their own. This requires the Aussie dollar to be substantially lower in order for our producers to compete internationally, particularly when many of the world’s central banks are already involved in depressing the value of their currencies. This outcome can be achieved by directly limiting asset sales foreign entities, or by intervening in foreign exchange markets. Yet neither of these two main options, which are used by other countries to great effect to manage their external position, seem to even be in the discussion.

The big mystery to me is why Glenn Stevens mentions these things now when he has had the power to intervene in currency markets in the interests of long-term Australian growth for a whole decade. Instead, he seems to be promoting a public debate that instead focusses on a magical, “one-sided coin solution” that is an accounting impossibility.

fn [1]. I also believe he uses the word asset to mean each of the two different types of capital.
fn [2]. I avoid incomes for the moment, as these are the result of previous international asset trades.


  1. This brings up an interesting issue Michael Hudson discusses in "A financial payments-flow analysis of US international transactions, 1960-1968": what goals do balance of payments accounting frameworks aim to accomplish? Are they meant to count every flow from one country to another or are they meant to assess the current financial position of the country involved?

    Hudson makes the case (which is interesting if not necessarily completely convincing) that imports, exports, interest payments etc should only be counted in balance of payments measures if they involve the transfer of a short maturity (one year or less) financial asset. This is under the assumption that balance of payments statistics should only count transactions that directly impact foreign exchange markets.

    " Even exports shipped on credit by U.S. firms to their own foreign affiliates (often for resale abroad) are treated as if these affiliates were provided with foreign exchange to pay for these goods, and are thus recorded both as export credits on current account and investment debits on capital account.About one
    third of reported U.S. credits on current account represent "wash" transactions such as these. In none of the above instances does current international payment take place: none involve any actual foreign exchange transaction"

    In other words, instead of

    value of biophysical exports-value of biophysical imports += trade surplus/deficit (abstracting from current accounts because there are complicated issues Hudson brings up regarding the treatment of financial flows and their relationship to real resource imports and exports)

    you use:

    short term net acquisition of financial assets-net accumulation of short term financial liabilities= balance of payments deficit/surplus (or "net-payment flow")

    In the latter framework, your example of physical machinery being transported to an affiliate/used to "pay for" an equity share of a project would not impact balance of payments positions directly. Given that you end your post with a discussion of financial policy and the relationship between the spot exchange rate and the current account, it seems a "net-payment flow" measure would be more directly useful for assessing different exchange rate policies and their effects.

  2. I read Stevens as using capital purely in the financial sense. He is saying that the imported financial capital can be used to finance physical capital buildup (useful) or it can be used for other purposes (not useful in his opinion). He is fine with the case you call "good foreign capital funded by bad foreign capital."

    1. But there is no such thing as "Foreign capital that builds new assets" in the financial sense. It is nonsense. Foreigner's can't just create new Australian assets in the financial sense. They always buy existing assets.

      Also, who decides what imports are "useful"? Almost every Australia import is physical capital - apart from a little food and fuel. Machines, vehicles, medical equipment, pipes etc. How can it not be useful?