insuranceadvisory

The academic community and financial market participants have finally come around to the idea that tracking capital or order flow in the developed world currencies was an important thing to do. Given the depth and liquidity of developed world markets, this fact may be surprising to some. However, as we have attempted to show above, this is nevertheless the case. Where it is true in the developed world markets that watching flow is an important pursuit in determining the exchange rate path, this is even more the case in the so-called emerging markets. Here, the relationship between individual capital flows and overall market liquidity is much more in favour of individual flows. Liquidity in emerging markets is by definition substantially less than in the developed world. Thus, individual flows can cause substantial price disruption in emerging markets, whereas they might be more easily absorbed in the developed markets. For this very reason, it is important for currency market participants who are involved in or exposed to emerging market currencies to have a reasonably accurate idea of the prevailing type of flows going through the market. One such is the EMFX Flow model (Robert Lustberg and Callum Henderson, 2001) created to track client flows going through the global dealing rooms of Citigroup in emerging market currencies. The type of information that one can glean from this is the following:
Total flow — The total flow indicator looks at cumulative transactions in the currency concerned against all the major base currencies (such as the Euro, dollar, yen, sterling and Swiss franc) combined. This gives an accurate indication of the client’s base total exposure to that currency.
Short- and medium-term flow indicators — The short-term flow indicator examines flow going through a specific exchange rate over the period of one month, while the medium- term flow indicator does this over six months.
Client-type flow — It is also useful to look at what types of clients are doing what. For instance, one can see whether or not corporate hedgers are being particularly active in a currency or not, or whether or not there is substantial speculative flow.
The use of such a flow model is to realize flow trends, in some cases confirming through the model what one knows anecdotally, and then make formal trading recommendations on the back of that. Some of the major finds in 2001 from this work were the following:
Brazilian real — Corporate hedging was the main flow dynamic of the dollar–real exchange rate for much of 2001. Examining the model’s findings we saw that the consistency of hedging activity, which entailed buying of dollars, and the quality of the types of corporations hedging, strongly suggested the dollar–real exchange rate would continue to appreciate. Equally, during November–December, when these corporations were no longer rolling their hedges it was no coincidence that the dollar–real exchange rate stabilized and retraced lower in favour of the real.
Mexican peso — Through the first eight months of 2001, client flows going through the dollar–peso exchange rate were largely in favour of the peso, confirming the anecdotal view that foreign direct investment remained a major supportive factor for the peso, more than offsetting the current account deficit. From October on, however, local corporations as a group turned net buyers. This did not cause an immediate appreciation in the dollar–peso exchange rate, but did put a floor under it and caused it to appreciate over time.
South African rand — There has been over the past few years much controversy regarding the flows going through the rand. Local market participants in South Africa have largely blamed the offshore market for rand weakness, while the economic community has been at a loss to explain rand weakness given the country’s “strong economic fundamentals”. From the EMFX Flow model, we discovered that the client flow of the bank had an asymmetric relationship with the price action of the dollar–rand exchange rate. That is to say, when clients were net sellers of dollar–rand — which was most of the time — the exchange rate remained largely range bound. On the other hand, during rare periods when clients became net buyers of dollars against rand, the exchange rate exploded higher. From this we can deduce a simple explanation for the rand’s weakness according to flow — locals are responsible for rand weakness.
Such findings can help greatly, not just in terms of providing trading recommendations for the speculative community, but in helping investors or corporations to plan their hedging strategies. Proprietary FX flow models focus generally on the short-term flow picture. However, there are flow reports that focus on the medium-term structural flows that go through asset markets. These also reflect useful information. As examples of such medium-term flow analysis, we now focus on the US Treasury’s “TIC” report and the Euro-zone capital flows report.