There has been a lot of flipping and flopping with interest rate and Aussie dollar forecasts through the course of 2014. I have been one of those changing views and forecasts when circumstances and data change.
The flow of news prompted me to look back at a long and slightly wonkish piece I wrote 3 years ago after a discussion I had with my mate Christopher Joye about economic forecasts and trading and the role of a bank economist.
It is a bit long, but please read it – it helps explain why getting a forecast wrong can still be profitable or getting a forecasts right can be next to useless if the forecast is already fully priced into the market.
From my old blog: 17 November 2011
Economists and Traders – Like Chalk and Cheese
Christopher Joye’s particularly interesting and well-reasoned article did a good job shedding some light on the similarities and differences between money market traders and market economists. His conclusions were largely based on an assumption that market economists and traders were trying to do a similar thing, that is, working out where markets were going. The difference was that traders take a shorter run perspective and economists or strategists look at the longer run and are required to produce “point” forecasts (e.g., the cash rate will be 3.25% in December 2012).
This is certainly correct for traders who only make money by buying low and selling high, often for a few basis points on a bond or half a cent move in the Aussie dollar. They do this through directional trades, options, spreads and a host of other market instruments.
But for market economists, such market moves are often irrelevant.
From my own experience as a market economist /strategist for around 15 years, most clients I saw were not interested in whether the RBA would cut or hike next month, wouldn’t care if after getting a CPI result you could confidently predict the RBA move in a few days time or whether a monthly set of jobs, retail sales or house price data were above or below expectations. Sure there was information in all of those events that often moved a market but any decent economist would hold a view through the noise of most events and only change that view when there was a material change in circumstances, events and perhaps market pricing.
For traders, the minute by minute news is critical. For market economists, it’s not.
Think of a game Test cricket as it evolves. Traders love a few quick wickets, a rapid fire century or a sporting declaration. Economists, on the other hand, like to look at the context of the game, knowing the pitch will crack on Day 5 and that rain is forecast tomorrow, meaning tactically, the batting team would have to be more aggressive. Based on that they predict the outcome noting the wickets falling and runs being scored along the way.
Economists are required to articulate a long run view based on point forecasts for the economy and markets a year or two out. In doing this, they are trying to develop a theme, a story, a narrative about where the key macroeconomic variables will go, and as a result, where market variables are likely to be.
They are sales people, story tellers, contemplating the future based on economic news and judgments about policy risks, event risks and they are even trying to judge where we are right now.
There is no doubt that point forecasts 1 year ahead are next to useless if that forecast is presented unconditionally.
When I was Chief Economist at Citibank in the late 1990s, I started to produce range forecasts – for example, I might have forecast that next months employment result will be between +5,000 and +10,000.This was saying that I expected only a moderate rise. Unfortunately, the wire services who collected these forecasts to generate the so-called “consensus” for each data release couldn’t cope with the range and invariably chose the mid-point. Traders didn’t understand either because for them, in the seconds after the data hit the screens, there was a big difference between 5,000 and 10,000. For me it was the same number and I would only be surprised with a negative result or one in excess of 20,000. That idea died a quiet death as a result.(I note in passing that the RBA produces range forecasts in its Quarterly Statements – so that’s progress.)
A current point forecast of mine – that the RBA will cut the cash rate to 3.5% by June 2012 – is really a manifestation of a view that the economy is slowing, inflation is set to fall sharply, the global economy is weak, fiscal policy will remain tight and unemployment will tick up. It is really not much more than this. If it turns out that the cash rate is 3.25% or 3.75%, it will be a pretty good forecast even though a 50 basis point variance for any trader is life or death.
That said, with the consensus on the cash rate at around 4.25% for June 2012, I am using my analytical skills to assess the above indicators differently from many others to suggest the consensus is in my view, going to be very wide of the mark. Those with a 4.25% forecast are using their analytical skills to implicitly say that my forecast will be wide of the mark. Fair enough – this makes the market work. I presume they have a very different view to me on inflation, GDP and maybe the global economy – we’ll see what happens.
Perhaps think of the process in terms of buying a stock for $1.00 and forecasting it to rise to $1.50 in one year. If it gets to $1.50 in a year, that is obviously a great call. Fantastic, profitable and a good investment. If it gets to $2.50, the $1.50 forecast is a big miss, horribly wrong and miles out… but any investor who bought the stock at $1.00 will be crying all the way to the bank. So too would it be the case if it got to $1.50 in three months.
See the story?
A critical reason why market economists need to produce longer run point forecasts is that they usually talk to long term investors. Those investors want to know medium term risks for investing in, say, Australia when compared with current market pricing. This is where some global knowledge and experience are important when presenting views, a point usually lacking from the experience of most traders.
The decision of a fund manager in Boston or Tokyo or Beijing or London or Muscat or Frankfurt to transact in the Australian market has little or nothing to do with the next RBA meeting or the next data set. It has nothing to do with a 10 basis point move in a bond or swap rate today or a two cent move in the Aussie dollar last week. Giving a view that you have the odds of an interest rate cut at 70% for next month is useless for a serious fund manager.
The decision of a fund manager to invest or not is based on, among other things, an assessment of the risks, return, relative value over the longer term.
A trader would have no idea how to articulate those views.
Economists and strategists can. For example, should a central bank or sovereign wealth fund in the Gulf States buy Australianbonds? Which ones? Government bonds? Semis? Which semi? A kangaroo bond? Which issue? Which tenure? Outright or on a spread? Should they receive swap? What about the AUD? Hedging is expensive so what are the risks the AUD falls to 0.80 next year? How would these markets look if the RBA did cut to 3.5%?
A trader would be left floundering trying to answer those questions because it’s not their job! Good strategists can.
Also, consider the example of the fund manager in Warsaw or Hong Kong who already has a long Australia position. They might want to know – what are the risks to my investment if the RBA cuts to 3.5% in June? What would it mean for my holding of X, Y and Z? They would boot you out of their office faster than you can say “bond trader” if you were to reply, “let’s wait and see the next CPI data point before I tell you” or “I’ll have more information after the RBA speech next week”.
That sort of approach is, from the position of a fund manger, sovereign wealth fund, central bank asset manager or real money manager, beyond contempt.
If they are to invest 2, 3 or 5% of their portfolio in Australia, they want to know well crafted long run views.
Traders and economists/strategists are chalk and cheese. They do different jobs. There is little if any overlap in the two functions and it is wrong to confuse them. That said, it is certainly possible to wear both hats – one can trade around the risks of the next RBA meeting for example, while holding a longer run view that the RBA will cut to 3.5% in June next year. These are entirely consistent positions.