Trading Tips

On Safe Havens and Currency Pair Correlations

I usually think the stuff put out by OANDA is pretty solid, but a post on their blog yesterday definitely fell short of that standard. Using their correlation matrix – which I do think is very useful and which I posted about on Facebook and Twitter not long ago – the author looked to make the case that USD/CHF is the best safe haven current pair.

First of all, currency pairs are not safe havens. That only works with individual currencies. At various points in recent years the USD, the CHF, and even the JPY have been tagged as safe haven currencies. This is an indication of where global capital is likely to move during times of stress. The result is that when the markets are risk-off, money will flow from risk currencies (currently the likes of the EUR and AUD) into safe haven currencies, and that will reverse with the markets go risk-on. This tends to make the pairs which join up safe haven and risk currencies (AUD/USD, for example) particularly volatile in times of shifting market psychology.

Shifting to the OANDA blog author’s argument, though, we get into the area of correlation. The point made is that USD/CHF is almost perfectly negatively correlated to EUR/USD in nearly all time frames. Let’s think about what those causes are:

1) The USD positions in the two pairs are opposing
Any two currency pairs which have the USD in them are going to be fairly strongly correlated most of the time. Whether that is negative or positive depends on whether the USD is in opposing positions in the pairs (base in one, quote in the other) or in the same position (base in both or quote in both). In the case of EUR/USD and USD/CHF we have the dollar as quote currency in one and base currency in the other, so we have a natural negative correlation.

2) The EUR and CHF tend to be influenced by common factors
This means both the euro and franc have a long history of seeing their values rise or fall on the basis of the same news and fundamental information. As a result, they tend to move in the same direction most of the time, though the amplitudes of the moves does vary (which is the cause for movement in EUR/CHF). Because EUR/USD and USD/CHF have the two currencies in opposing positions (as noted with the dollar above), there is a natural negative correlation between the pairs.

3) The Swiss National Bank has put a floor under EUR/CHF at 1.20
Because of the persistent issues in the Eurozone and the perception of the franc as a safe haven (meaning flows out of the EUR and into the CHF), this has effectively mean EUR/CHF has been pegged at or near 1.20 for long period of time of late. That means the euro and franc have traded in near lockstep. Going back to the relationship between EUR/USD and USD/CHF noted in 2) above, this is the strongest case of all for a very strongly negative (near -1) correlation between the two pairs.

Put all this together and you get the reason for USD/CHF trading at a near perfect negative correlation to EUR/USD being that the EUR and CHF are virtually the same currency these days, and both pairs feature the same opposing currency (USD) on the other side. Of course they are going to be strongly negatively correlated!

Because of the strong negative correlation between EUR/USD and USD/CHF, hedgers may look at the two as presenting an opportunity. I would note, however, that being long EUR/USD (for example) and hedging with a long USD/CHF position would only serve in creating a synthetic (and overly expensive) EUR/CHF long position.

The bottom line with the correlation stuff, which seems to get everyone very excited these days for some reason, is that it’s important to know why two markets are correlated (or uncorrelated). If you don’t, then you may find yourself in a nasty position where the correlation breaks down and catches you out because you weren’t prepared for it based on an understanding of the market dynamics at work.

And getting back to the idea of USD/CHF being the best safe haven currency pair… I’m sorry, but just looking at the correlation to EUR/USD alone is a far cry from sufficient evidence to make such a claim. You’d have to do a full analysis to see which pair or pairs is most reactive to safe haven capital flows before making any claims. Correlation analysis doesn’t get you there.

Trading Tips

Using Secondary Indications in Your Market Analysis

Yesterday Adam at Forex Blog put up a blog post looking at the British Pound, specifically in terms of GBP/USD. He throws a lot of different stuff into his assessment of the UK currency, part of which is looking at the prospects for a rate hike by the Bank of England. To that end, let me share two charts I keep an eye on in my work.

This first chart shows the spread of UK 2 year Gilt rates over the US 2yr Treasury Note rate, with the spread’s correlation to GBP/USD in red as the lower plot.

The second chart is the same as the one above, but swapping German Bunds in place of US Treasuries and running the correlation against EUR/GBP instead of GBP/USD.

I offer up these charts for a couple of reasons. One is to show the sort of secondary analysis professionals use to assess the markets. Another is to show how frequently market correlations can change. We would expect a positive linkage between the UK/US rate spread and GBP/USD and a negative one between the UK/German spread and EUR/GBP, but that’s not always the case.

The third reason for showing these charts is to show what’s been going on in these spreads lately. The UK/German spread has fallen sharply, strongly indicating the market’s view on whether it will be the ECB or BoE the moves first to hike rates has moved strongly in favor of the former. Things are less dramatic in the UK/US spread, but the breakdown there hints that fixed income traders have become less confident about a BoE rate move in the short-term in general, not just as opposed to the timing of the ECBs action. These are the sorts of things the professionals are looking at and thinking about in making their market judgements. It’s all related.

Reader Questions Answered

Correlation analysis for trading position diversification

I got the following inquiry from the ever-curious Trader Rod.

A few questions concerning correlation between two instruments for the purpose of diversification:

1. When calculating the correlation coefficient, given my time frame is daily, should I focus on daily closing prices or daily returns?

2. If the answer to previous question is to focus on returns, and the average holding period for a position is one week, should I calculate the coefficient using series of weekly returns?

3. What is a good threshold for diversification, i.e. at which point do I decide the two instruments are too highly correlated, e.g. anything below 0.7 is acceptable ?

Correlations have been gaining in focus of late, no doubt in large part due to the very obvious and public linkages between the dollar and stocks. It’s hardly a new thing, though. It’s been used in the type of work Rod is describing here for decades. Modern Portfolio Theory, for example, is heavily involved with correlation analysis. In other words, this isn’t a new field of analysis, so if it’s something you have a mind to explore there’s plenty of reading you can find on the subject. Make sure you do take a half approach, though. Using correlations without understanding what they are, what they mean, and how they can change can be extremely dangerous. Just ask the blown up banks and hedge funds who didn’t make it through the credit crisis.

As for Rod’s specific questions:

1) You should always use returns. They are how you can compare two instruments in a standardized fashion. Otherwise, differences in the prices of the securities can create comparison problems.

2) A holding period of only a week makes correlation application challenging. They tend not to be that stable in the short time frames. At best I’d say look at a rolling 5-week correlation and use it as a rough guideline. But realize that it’s only rough. There’s likely to be considerable deviation.

3) To achieve true diversification between two securities you need to have a near zero or negative correlation.

Reader Questions Answered

The Relationship Between Stocks and Forex

I was asked a question by someone on my mailing list yesterday about the relationship between the forex and stock markets. Normally I include the text of the person’s question, but in this case English is not the person’s first language and the inquiry is fairly lengthy, so I’ll skip that in this particular instance. The bottom line is that like many folks this individual is wondering about why the dollar falls when stocks rally, and vice versa.

Following the Patterns and Relationships
As a professional forex market analyst I spend my days watching the all markets (not just forex) and seeing the various interactions between them. Some are subtle, fleeting, and likely only to be spotted by those looking specifically for them. Others are broad and obvious that anyone can see. The whole stocks up/dollar down pattern we’ve seen for some time now is definitely in the latter camp. Even the media is aware of it. 🙂

Back in June I did a presentation on cross-market analysis and trading at the L.A. Traders Expo. I was asked the question then what market tends to lead the rest. My answer was that it varies from day to day, and sometimes even hour to hour (for example the bond market has been tipping off dollar moves the last couple days), but in very general terms it tends to be the forex market which moves first, fixed income (interest rates) which follows second, and equities bringing up the rear (commodities probably falls in the fixed income area on the timeline, but they vary).

Risk Aversion and Carry Trade
The relationships between the markets varies considerably, though, so don’t make the assumption that what’s going on now will persist. Right now we have a situation where a couple of things are contributing to the inverse relationship between the dollar and stocks. One is the general risk aversion/flight to quality trade in which nervous market participants will tend to move money into the dollar for safety.  This isn’t as dominant a theme as it was a few months back, but is still a factor.

The other is the fact that the dollar has to a certain degree supplanted the yen as the carry trade short currency of choice (meaning the one that gets borrowed to be converted into a higher interest bearing currency). The carry trade is a highly risk sensitive position. Those trades will be closed when people are worried about the global economy partly on a safety basis and partly on the view that the higher rates could come down.

Seasonal Patterns in Action
Also keep in mind that so called “seasonal” patterns can influence things. This time of year has generally been a bearish one for stocks, but it has also been that way for the dollar (see Opportunities in Forex Calendar Trading Patterns). That creates an interesting conflict between the usual patterns. So far in September the dollar pattern, because of the inverse relationship, has cancelled out the usual bearish action we’d expect in stocks this month.

Trading Tips

Why are stocks and the dollar trading inversely?

One of the questions that’s come up a few times of late on the discussion boards is why the US stock market and the dollar are trading in the opposite direction from each other. This has actually been going on for a while now. No doubt folks who have picked up on the pattern are trying to think about how they can use it in their trading.

Here’s the deal.

Stocks and the dollar are mostly moving on the same trigger at the moment, the risk aversion one. When the market is scared they sell stocks and buy the dollar as a flight to safety trade. That usually also means buying Treasury debt, which is why stocks and Treasury yields have been moving in the same direction for the most part. When the market is feeling more happy and the fear has been put aside, stocks rally and money moves out of Treasury securities and the dollar.

Do not consider this any kind of permanent situation. Correlations in the markets come and go all the time. As many a quant trader found out in this cycle, even markets that are normally uncorrelated can become very correlated. It’s not a good idea to rely on these things continuing on perpetually.