26% GDP Growth Or Bad Accounting? - Bezek's Daily Briefing

by: Ian Bezek


Tuesday appears to be the big "risk-on" day for global markets.

Irish GDP grew at 26% in 2015. What to make of such an outlier statistic.

How I judge economic statistics as it comes to my thinking toward investments.

Stocks moved higher on Monday, though the move lacked conviction. The S&P 500 (NYSEARCA:SPY) made new highs but didn't actually advance that far, particularly after the late day sell-off into the close.

Tuesday appears to be the real "risk-on" day that will get the market clearly into uncharted territory.

For this rally to make substantial forward progress, we needed to see several of the recent things that had been bothering the market firm up. That appears to be happening now at last.

To that end, we have the British pound (NYSEARCA:FXB) sharply higher today, up almost 2 cents to $1.32. The run under $1.30 had set the stage for much bigger problems, but today's move seems to remove that risk for the time being. The Japanese yen (NYSEARCA:FXY) has tanked in recent days after coming right up to the pivotal 100 level against the dollar that had global macro traders extremely nervous.

Crude oil (NYSEARCA:USO) is up more than a buck today to get back over $45. Oil is still struggling and is way off the recent highs, but merely stabilizing above $45 should be enough to allow the bull movement on equities to continue.

Treasury bonds (NYSEARCA:TLT) are (finally) plunging. I'd been waiting for this for a week now; it was deeply troubling how quickly yields were sinking, even with stocks rebounding. This is the first "normal" day of action we've had out of bonds since Brexit. The big panic in the credit markets may finally be lifting - technically, TLT is now in danger with that top being a striking reversal pattern:

Finally gold (NYSEARCA:GLD) is off more than $10/oz today, building on its recent softness since peaking a week ago Tuesday. However, gold is still in rarefied air up here around $1,345; with Brexit being ignored by the market for now, there's really no reason most momentum traders would choose to stick with a safe haven trade at this point. Assuming the rally in the S&P 500 has any legs, we could see a big break in gold and related mining equities (NYSEARCA:GDX) in coming days.

Misleading Statistics

My university training was in economics; not finance or business as you may have expected for someone that ended up working at a hedge fund. And let me tell you, my opinion of economics significantly declined during my four years pursuing my degree.

It's no mistake that economics, at most universities is a "liberal art" and not a "science." While the Nobel committee may give out a prize for economics, there's a great deal our models and calculations still can't come close to figuring out.

One particular blow for my faith in economics came during my econometrics class where we learned advanced methods for trying to find statistical correlation within large pools of data with multiple potential causes. While we were supposed to build realistic models that demonstrated economic cause and effect, I threw a bunch of random data into my program to see if I could tease "causality" out of gibberish.

At one point, I managed to arrive at a multi-factor model that seemed to prove that US carrot production and the number of wins by the Denver Broncos football team in a season were predictive of the US inflation rate to a 95% degree of confidence. In other words: garbage in, garbage out - if you tinker enough, you can make models say just about anything.

I'm reminded of this upon news that Ireland's GDP "grew" at 26% in 2015. This final tally is a rather dramatic difference from the preliminary growth estimate for 2015, which had been seen at a robust but far lesser 8% rate. Where'd the other 18% GDP growth suddenly show up from? And should we be flinging money at Irish stocks (NYSEARCA:EIRL) to celebrate the news?

The answer to the last question is probably a no. It turns out that as the Irish statistic office reviewed the data, large changes in aircraft purchases, a surge in foreign trade, and the arrival of new assets to Ireland from buyouts that moved large multinationals' headquarters to Ireland to engage in tax arbitrage caused the huge jump in GDP.

None of these factors are of particular assistance to the average Irish person, and it's fair to say median income didn't rise anything remotely close to 26%, despite the sudden reported surge in Irish GDP per capita.

The "GDP growth" caused by buyouts that left companies nominally headquartered in Ireland but with most of operations elsewhere is particularly interesting. As companies become more interested in sophisticated tax arbitrage, will we see more such one-time surges (and plunges) in GDPs of countries? Particularly for small countries that serve as tax havens or finance centers, large corporate efforts to game the tax code could end up leaving GDP data very noisy and relatively useless for economic forecasting.

Of interest is the impact on the bond market. Ireland's Debt/GDP ratio plunges to below 80% with this new figure, down from a much more troubling 123% figure as recently as two years ago. If you're a creditor (or simply buying government bonds), be very careful that you don't get a false sense of security from a ratio that suddenly makes the debtor appear far more solvent.

If Ireland's economy is constructed in such a way that GDP can rise more than 25% one year, what's to stop it from experiencing a sudden plunge if tax loopholes close or that surge in foreign trade reverses?

The extreme Irish GDP print is just one example of a much wider point. Beware of making your investing decisions based on statistics in isolation, particularly if the data is particularly volatile. It's easy to build a compelling case if you select your statistics with bias. Similarly, backtested investing models rarely play out so well in real market conditions, generally because the backtester has sifted through many things that didn't work to find one that (probably by accident) did.

Finally, be sure to ask yourself, does this statistic make sense? Back when inflation was a more popular subject matter, I remember being frequently confronted by people citing Shadowstats "alternate" inflation statistics. Shadowstats claims that, for whatever reason, the US government understates inflation by a massive margin - many percent annually.

According to Shadowstats, consumer inflation has averaged roughly 10% in the US since 2000. In 2008, it supposedly topped out at 13%; for those of you who lived through the 1970s, you know 2008 in no way resembled that stagflationary era. If not, ask your parents. Finally, simple math using the law of 72 suggests that a 10% inflation rate means prices double every 7 years.

Thus: Since 2002, according to Shadowstats, prices have doubled and then doubled again. Think about your spending now: does the average good cost more than 4x as much as it did 15 years ago? Of course not. Inflation is a real thing, but it's not occurring anywhere near that rate that this "alternate" metric claims.

Before buying gold from a fear-based presentation claiming the US is suffering from runaway inflation, for example, spend a couple minutes thinking about the statistics used. If they don't seem plausible, they've probably been cooked up.

Back to Irish GDP, the statistics department there hasn't (it would seem) done any of its math wrong or with an intention to mislead. But the net result, that the GDP figure claims Ireland is experiencing some sort of historic record-breaking boom, is totally bunk.

As Investec economist Philip O'Sullivan put it: "The new headline growth figures for Ireland bear scant relationship to developments on the ground here." I couldn't have phrased it better myself.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.