The author has been described by News Ltd as an "iconoclast", "Svengali", a pollie's "economist muse", and "pungently accurate". Fairfax says he is a "Renaissance man" and "one of Australia’s most respected analysts." Stephen Koukoulas concludes that he is "85% right", and "would make a great Opposition leader." Terry McCrann claims the author thinks "‘nuance’ is a trendy village in the south of France", but can be "scintillating" when he thinks "clearly". The ACTU reckons he’s "an enigma wrapped in a Bloomberg terminal, wrapped in some apparently well-honed abs."

Sunday, May 13, 2012

The RBA's inside story

In an interview a few weeks ago, Sky's Peter Switzer kindly introduced me as a "great friend of the RBA." In light of the many criticisms I have aired over the years, I couldn't help but laugh at this characterisation. I subsequently mentioned it to an RBA executive. Almost as amusingly, this person retorted: "yeah, 60% of the time."

Unfortunately, today may be one of those 'forty percenters'. What I am going to write about you will likely never read in an Australian newspaper. The journalists in question cannot report on these issues for fear of breaching confidences given to their sources.

Because I have one foot directly in financial services, and another in the media as a public commentator, I am a little different. Unlike most other RBA watchers, I never speak to the Bank, its board, or its political masters about monetary policy. I don't get the benefit--or, as we shall see, suffer the costs--of receiving the all-important private telephone and in-person briefings that senior correspondents secure after important RBA statements, and before major decisions.

I do speak to the Bank about abstract policy issues, such as whether they should use interest rates to lean against asset prices, or technical data matters, like how house prices movements should be measured.

I've also asked the Bank to set me straight if I ever make an objective mis-step. As one example, the Bank has a habit of frequently emphasising the "mid-point" of its 2% to 3% per annum inflation 'target' as its policy goal. I've previously written that this can lead one to infer that the RBA's 'implied' inflation target is actually a more specific 2.5 per cent. Core inflation outcomes materially above or below this point estimate could, therefore, be judged as misses.

It was, however, pointed out to me that Guy Debelle and Glenn Stevens had published a paper in 1995 arguing that the RBA's monetary policy aim is a deliberately "thick point", or wide, band most accurately described as "two point something". It is not an artificially precise 2.5%.

Long story short, anywhere in the band satisfies the inflation objective because the RBA knows that it cannot over-engineer its inflation targeting outcomes--the standard errors associated with its forecasts are just too high. That, by the way, is one reason why there is nothing wrong with consistently punching out 2% per annum core inflation, as is currently the case. (Most developed world central banks consider 2% normal.)

The motivation for today's remarks derive from questions I received last Friday. A senior interest rate trader asked me what I thought the RBA's closely-read Statement on Monetary Policy might say (it was to be published that day). I felt the lethargy mount inside me before I mustered a shallow response. All I could offer was, "dovish mate--they'v gotta rationalise the 50 points, ex post facto." But then I quickly added that I had not thought about it much.

After the Statement came out--which was dovish--Aussie government bond prices jumped a bit. Another experienced market participant called me and queried my views. I said I hadn't even read it aside from a quick glance at the housing section, which was in line with my priors (ie, the RBA is pushing a stabilisation meme, now couched as a deceleration in house price losses). I said I'd get back to him when I had some thoughts (I still don't).

Over the weekend I started reflecting on why I had so little interest in what is supposed to be the RBA's premier publication. It slowly dawned on me: it was more-or-less meaningless for rates. It would be part exercise in justification of a decision the RBA staff did not recommend, and part hazy and contradictory glimpses of their view of the world.

Rather than telescope in on a central set of projections that the staff actually believed in, the chastened executives would likely be obscuring any sharper messages with insurance and obligatory acknowledgements of how they had purportedly 'misread' the economy. (Yesterday's 4.9% unemployment print combined with the low core inflation over the last three quarters tells us that the RBA had, in fact, set monetary policy almost perfectly.)

As it transpired, the RBA did indeed make its views even fuzzier in this particular Statement by more-or-less giving up on forecasting. The projections the RBA delivered for year-ended GDP growth and core inflation were so wide that any real-world results at the limits of these ranges could have diametrically different consequences for policy. Beyond a year or so, the RBA has now formally embraced Glenn Stevens's 'first rule of forecasting': don't forecast; or, if you do have to forecast, hedge your bets.

My main idea before the last RBA board meeting was that no matter what happened we would learn a lot about how the Bank thinks and operates. You see, after the stunningly low first quarter inflation numbers, the RBA's 50 basis point cut was not a surprise, per se.

Before the inflation data, I had published a 'cheat sheet' with a table claiming that if core inflation came out at 0.3-0.4% one could expect the RBA to cut by 25-50 points. I also quickly attached a 40% probability to 50 basis points after the release. And by the time the Board meeting came around, the financial markets had converged to roughly this same estimate.

To be sure, 50 basis points before the CPI release was very low probability. Only one person in Australia was pulling for it: Stephen Koukoulas (or "Koukie" to friends). And with the benefit of hindsight, Koukie made a majestically timed call. The only wrinkle is that he had also called 25 and 50 basis point cuts in February and April too. So May was more about getting the medium term right, although this week's unemployment data arguably call this into question.

The surprising thing about the RBA's May meeting was that the senior staff only recommended 25 and went to considerable lengths to resist a larger cut. The surprise was that the once all-controlling RBA executive had again had their central recommendation--which would have been strategically hedged--rejected by the RBA's dovish, and increasingly powerful, board.

That's what happens when the "insiders" hold only two of the nine board seats, stakeholders with a vested interest in lower rates dominate all the others, and when the Governor is an open-minded consensus builder who has hitched his legacy to enhancing governance and transparency.

On the morning of the board meeting, a very senior economics correspondent for Fairfax, Peter Martin, published the following remarks in the Sydney Morning Herald:

"The staff recommendation sent to board members on Friday is believed to recommend a cut of 25 points, arguing that the economy is not weak enough to demand anything larger and that rates can be cut again if needed."

As a seasoned journalist with impeccable RBA and Treasury contacts, Martin would never make this statement unless he believed it to be true on the basis of a primary source. The day before, David Uren, another senior and instinctively cautious senior economics correspondent with The Australian, offered these predictions on both what would happen at the meeting, and, more curiously, what specific GDP growth numbers the RBA's Statement on Monetary Policy would include:

"People speculating that the Reserve Bank may cut by 50 basis points tomorrow are hyperventilating. The downgrade in the bank's growth forecast is unlikely to be dramatic. National Australia Bank's business survey shows conditions are still in line with the long-term trend and unemployment has remained steady for about two years. The March quarter inflation report was soft, but there is nothing to suggest that the economy is entering a recession. The Reserve Bank's monetary policy statement, which will be published on Friday, will still show growth over 2012-13 of at least 3 per cent. The bank will not be badgered into repeated slashing of rates by those parts of the economy that are on the wrong side of the structural change which the high value of the Australian dollar is forcing. It is unlikely that the bank judges that its cash rate is far from where it needs to be."

Like Martin, Uren would likely never write this and, more pointedly, dismiss the prospect of a 50 basis point cut, unless he had been guided by the RBA staff to do so. While he obviously got the rate call wrong, he was, perhaps unsurprisingly, spot-on with his even more difficult-to-anticipate Statement on Monetary Policy forecasts. Of course, the RBA staff determine the numbers included in the Statement without interference from the board in contrast to the cash rate decision.

Just as the staff are inclined to advocate their positions via preferred media proxies (universally accepted as a common, if awkward, practice), it's possible that some of the seven remaining board members do so too. This may be the reason why News Limited's leading columnist, Terry McCrann, (confidently) missed the RBA's February and April decisions to pause. At both meetings, the staff evidently managed to hold the line against a board agitating for rate relief.

But in May, McCrann was uncannily on-the-money well before any other analyst, or the market, save  Koukoulas. On April 25th, after the CPI report but a week before the board meeting, Terry called a 50 basis point cut yet went one step further to, remarkably, describe a standard cut as a "most unlikely circumstance." Terry's smart, but he's not that smart. I am guessing that the reason he could effectively rule-out 25 is because either someone on the board (likely), or someone at the Bank (unlikely), told him it wasn't going to happen:

"TWO things are now absolutely beyond doubt. There will be an official interest rate cut next Tuesday, and the official rate will be cut by at least 50 points over the next five weeks. Somewhat less certain, is when the 50 points will be delivered. I suggest in one hit on Tuesday. But in the most unlikely circumstance that Tuesday turned out to be only 25 points, that would simply make a further 25 points absolutely certain at the next Reserve Bank board meeting in June."

On April 29 McCrann affirmed this projection:

"ON Tuesday, the Reserve Bank will cut its official interest rate...It should start with a 50-point cut and I suggest there's a better than even chance it will."

Now just think for a moment. If it is possible for this information to leak out from the RBA's board, and from senior staff, is it conceivable that there are folks around the world who might seek to profit from it? I would venture so. Indeed, there was some rather unusual trading on the day of the RBA's last meeting.

At exactly the same time the RBA board meeting finished--at 12:30pm--through until the time the policy decision was announced at 2.30pm, the Aussie dollar commenced a slide from circa 1.0422 US cents to as low as 1.0388 cents. There was also a slightly odd spike in 3 year government bond prices, which directionally has the same meaning (Koukoulas tweeted about it), although they pulled back on low volumes just before the release. Maybe it was noise. Maybe not. I honestly don't know.

Bruce Kovner, the founder of one of the world's biggest and most successful hedge funds, Caxton Associates, has said that in the 1980s the best traders of currencies were the Russians because of the KGB's ability to pick up economic intelligence. We have many private companies with intelligence gathering capabilities, and foreign intelligence agencies, functioning in Australia today. It is well documented that sourcing non-public economic intelligence, and pro-active industrial espionage, are key aims of any sovereign intelligence entity, including Australia's. I am not saying anything more than that.

As a hypothetical, you've got a full two hours between the 2:30pm release date and the time when seven of the nine RBA board members drift out from its Martin Place headquarters to monitor communications and collect one solid indication of what might have happened. The RBA takes precautions, of course, including sweeping its board-room for bugs.  But if you are successful, you could make tens, if not hundreds, of millions of dollars. It is hardly unimanginable. The technology certainly exists. The self-interest is a given.

With this in mind, I would note that Dow Jones's Enda Curran highlighted this ABS media release during the week revealing that the surprisingly strong, and price sensitive, retail trade data had been leaked to six clients 20 minutes before it was communicated to the market. One assumes this was just a technical glitch.

Any long-time policy analyst knows that there is more than enough soft-corruption, influence-peddling, and subjugation of taxpayers' interests embedded in Australia's public sector decision-making processes for one to be worried. I have regularly hammered on about the wealth-destroying biases ingrained in our saving system's asset-allocation decisions and the manifold unpriced taxpayer subsidises bequeathed to our too-big-to-fail banks. Yet remedies appear hard to implement when even the nation's political leaders argue in private that the bureaucracy is pathologically resistant to change.