This page has been archived and commenting is disabled.
Goldman Launches A Macroeconomic "Surprise" Tracker, Finds Nothing But Disappointment
Goldman's macroeconomists led by Jan Hatzius have launched a new tool, the US-MAP, also known as a Macro-data Assessment Platform, whose purposes will be to track up/downside macroeconomic "surprises" relative to expectations. As lately we have had nothing but downside surprises, yet the market keeps ploughing higher for some computerized reason, this would be a useful tool to keep track of just how overoptimistic the economic consensus is. Yet as usual, Goldman introduces a fudge factor in the way it grades the importance of specific economic data: for example according to Goldman's relevance matrix, such lagging and manipulated data as the Conference Board Confidence Index is given the same weight as the ISM Services and the ADP report, and greater weight than Initial Claims, New, Existing and Pending Home Sales, the Empire Index, Industrial Production, Core PPI , the Trade Balance, Housing Starts and, stunningly, actual Consumer Spending. How this makes any sense, we don't know - then again we did not win the Wall Street All Star Economic Prize last year, so address your queries to Jan. Yet even with all this doctoring, Hatzius is forced to admit: "A visual check of US-MAP readings for the data released thus far in June confirms the predominance of downside surprises." In other words, it is about time economists start factoring in a double dip instead of continuing to live in the land of unicorns, endless stimulus and pixie dust. The game of "the data beat expectations" is now over. Now what?
Below is a chart of Goldman's incident and cumulative downside surprise readings:
As for the doctoring of the data, here is how Goldman explains its multidimension matrix for assigning data relevancy:
The key insight of GS-MAP is to recognize that most questions about economic reports turn on one of two issues: (1) how important, or relevant, is the indicator being reported, and (2) how surprising is the outcome in the particular report that has just been issued? For the rating systems that have been introduced thus far, GS-MAP works like this:
- A relevance score is assigned to each indicator, based on its correlation to real GDP growth in the associated country or region. For simplicity, the score runs from 0 to 5, with 5 identifying variables with the highest growth correlations. It does not change from one month to the next.
- A surprise score measures the extent to which the outcome of a particular report exceeds or falls short of expectations. It is based on the difference between the reported figure and the "consensus" expectation. Again, the score runs from 0 to 5 in absolute value, providing a total range of -5 to +5.
A composite GS-MAP score is then calculated as the product of the relevance and surprise scores and presented as a gauge of the overall importance of the data that have just been released. However, the two components are also reported; thus, the full GS-MAP reading for an "extreme" downside surprise in an indicator of "medium" relevance would be -15 (3,-5).
For US-MAP, we will follow the same general approach of assigning relevance and surprise scores for 28 indicators of economic activity and inflation, but with two important changes:
1. The relevance score will be based on market impact rather than on growth correlations. Specifically, we have updated and expanded the analysis used in the February 2007 revamping of the Goldman Sachs Surprise Index (GSSI), in which we estimated the effect of a normalized deviation from the median forecast (as reported by Bloomberg) on intraday movements in the yield on 2-year Treasury notes. The update adds more than three years of data to the estimates derived at that time; the expansion adds 12 new variables to the 16 already included in the GSSI. (We have yet to change weights in the GSSI, pending decisions on which of the new indicators exert effects large enough to warrant inclusion and which of the old ones might be dropped.)
This departure from other applications of GS-MAP reflects three considerations that we believe are unique to how financial markets process US economic data. First, the data are more extensive than for most other countries, and many market participants are reasonably familiar with how the data fit together. Hence, the information content of a relevance score based on correlations with real GDP growth is much lower for the United States than it is for other countries. In fact, this familiarity is such that US reports often have a much larger impact on trading in foreign markets than do the reports on local economic activity.
Second, a large body of data has developed on the outcomes of many US economic reports compared to expectations. Thus, it is possible to estimate the relative importance of these reports based on how surprises have affected markets in the past. This provides a more natural link between the two dimensions of US-MAP and, we think, a more appropriate measure of relevance.
The difference between relevance as measured by market impact and relevance as measured by the (concurrent) correlation with GDP growth is not trivial, as shown in Exhibit 1. Over the past ten years or so, the correlation between the two is only 0.22, indicating that markets trade on more than what indicators say about economic growth.
Thus the third reason: simple growth correlations yield results that sometimes are at odds with impressions of relevance that are admittedly more intuitive but nonetheless valid in our view. The most glaring of these is that inflation indicators often exhibit low correlations with real GDP growth, yet they are quite relevant to the markets. The core CPI illustrates the point as it has been virtually uncorrelated with real GDP growth over the past decade. At the other end of the spectrum, a score based on growth correlations would put the Empire index of manufacturing conditions in New York State at the top of the list. While we are prepared to accept the verdict that this indicator is a better proxy for real GDP growth than we may have realized, it is nonetheless hard to believe that it outranks payrolls, the ISM index, and retail sales (and everything else) in relevance.
2. We may make judgmental adjustments to the surprise score. This could be more controversial than our modification of the relevance score, as it introduces the possibility of ad hoc intervention. However, there are times when a large surprise is due to factors of dubious significance and other times when a result close to expectations masks other data of importance; this is particularly a risk when reports are large and complex. The labor market report, with the potential for a contrary move in the high-profile jobless rate, is the best example of this. In such situations, we see it as our job to inform clients of inconsistencies, which we already do in our written commentary. And since this is the case, we should do likewise in the surprise score of US-MAP.
Last Friday's retail sales report for May shows the need for this flexibility. The median forecast for sales of goods other than autos was for a 0.1% increase, but a 1.1% decline was reported. The difference was more than 2 standard deviations. However, this surprise was mostly due to a tax related correction in the building materials component, which in any event is discarded by the Commerce Department when it calculates personal spending. Under these circumstances, it would not make sense to stick with a high surprise score. Likewise, the Philadelphia Fed index for June fell sharply, but the detail in this report did not support such a large move. Although examples like this abound, we intend to be sparing in making adjustments, especially large ones. It is also worth noting that "unlike our market-oriented change in how the relevance score is determined" judgmental adjustments to the surprise score will convey more information about economic significance, at least as we see it, than about market significance.
And here is how the two-dimensional matrix of the various economic measurements would look like in real life. Remember: the greater the relevance, the more important it is.
At this point it probably makes sense to recall that wonderfull essay "Understanding Poetry" by Dr. J. Evans Pritchard, Ph.D., quoted in Dead Poet's Society.
'Understanding Poetry,' by Dr. J. Evans Pritchard, Ph.D.
To fully understand poetry, we must first be fluent with its meter, rhyme and figures of speech, then ask two questions: 1) How artfully has the objective of the poem been rendered and 2) How important is that objective? Question 1 rates the poem's perfection; question 2 rates its importance. And once these questions have been answered, determining the poem's greatness becomes a relatively simple matter.
If the poem's score for perfection is plotted on the horizontal of a graph and its importance is plotted on the vertical, then calculating the total area of the poem yields the measure of its greatness.
A sonnet by Byron might score high on the vertical but only average on the horizontal. A Shakespearean sonnet, on the other hand, would score high both horizontally and vertically, yielding a massive total area, thereby revealing the poem to be truly great. As you proceed through the poetry in this book, practice this rating method. As your ability to evaluate poems in this matter grows, so will, so will your enjoyment and understanding of poetry.
Dr. J. Hatzius, Ph.D. has merely taken this essay and converted it into one entitled "Understanding and Mitigating Downside Economic Surprises." Next up: writing the most persuasive haiku to describe the plunge in NFPs.
We wonder how long before the snake oil is put away and the economist community realizes that absent another multi-trillions fiscal and monetary stimulus package, the double dip (or the accelerating in the slide, depending on how one views these things) is unavoidable. June was the first month in which the cumulative negative surprises came far worse than expectations (not like stocks care of course). And if you thought June was bad, wait until July, August, September, October, etc, etc. and so forth until we eventually hit 2020 when US Debt (pro forma for all off balance sheet insanity) to GDP will be a four digit number.
»
- 10772 reads
- Printer-friendly version
- Send to friend
- advertisements -
FOG MAP.
Not. Have any of you douches asked yourself how the fuck 'Tyler' knows about this Tracker and the results?
Goldman == ZH
Welcome to the world of disinformation.
Don't forget: ZH also = Fed, = Bilderberg, = LBMA, = CIA
Has anyone told him about the secret rings we get upon being registered w/ ZH for a full year?
Forget the rings. Since (apparently) we're all douches, I would rather have some Summer's Eve for Men.
Not only do I want to be well informed, I want to stay fresh all day. That's all.
http://www.summerseve.com/Feminine-Powder.php
Ah shut up, all of you spin doctors. I know who you are. You're not as clever as you think.
Your forgot to add CFR and Committee of 300 to that list.
They ought to team up with Google and
provide GNP tracking by the minute. Gross National Product displayed by the minute, in NY times sqaure, with the help of Googles , cafe IP address sniffin, moble video cams/,
>> Maybe google can issue a national credit card and back it with ??
All that electricty that they will need to farm those machines to solve your complex problems (or just record every key stroke you make)
Ay yes, an SDR backed by electricity?
so, ah, where does the discovery of "trillions of dollars" worth of precious metals in afghanistan fit on this scale?
rolling.
That's a ruse to get Japan's bravest or most naive geologists kidnapped by the Taliban.
Wasn't knowing more and more about less and less what brought LTCM to its knees?
DavidC
The revenge of fuzzy math!
Now what? - The word "unexpected" gets used more than vowels
Like "I just shat my pants - unexpected like"
I went to the bath house. Unexpected butt seck!
Ta da! I'll be here all week.
Holy shit,,there I am.
"Normalized Surprise" -- sounds like a breakfast cereal!
Who the fuck pays attention to any new index or idicator? All of them are manipulated and adjusted for something, not to mention the data they use is already adjusted for something, time of the year, odd or even year, hopes of government, mood of analysts, consumer habits, hedonistic adjustment and so on!
I love standards, there are so many of them.
- Engineering principal #7485
Wake me up when the major Wall Street banks start losing money. Then, and only then, will the great unraveling commence again.
Unemployment? Who cares? PIGS implode? Who cares? Debts and deficits? Who cares? The ocean is being turned to roofing tar? Who cares?
The market may drift here and there, but will remain generally buoyant until the major banks begin to loose money again. It will go up when it should go down; or down when it should go up; zig when it should zag--all because the market is programmed to do maximum damage on a millisecond by millisecond scale. Good luck trading in that environment.
Until then, all data is just easily manipulated noise.
Remember what happened on March 9th that reversed the plunge? Citi was rumored to be profitable.
Since then, not a shred of negative news has mattered that wasn't forgotten a few weeks later. I can only suspect that this trend of "nothing really matters" will continue for quite some time.
Nothing really matters ...
Any-one can see ...
Nothing really matters ...
... to meeeeeeeeeeee.
Mama just killed a man...<next>
thumbs up
ah yes, you woke up, then went back to sleep...that's intellectual property that i own dude.
You obviously don't get it Tyler.
In this brave new world (bnw) economics measuring what people do is worthless!
What really matters is what people expect others will do,
especially what they expect others will do for them.
Of course the Conference Board Confidence Index and the ISM Services indexes counts higher than stupid numbers reflecting stupid peoples actions such as the initial claims, pending home sales and consumer spending.
Now please get up to date with reality as it is depicted to us.
Coming Live to You from Our Flag Ship Station in the "Get in Step" Network we bring you and hour with you favorite commentator Tyler Durden. Tyler burst back onto the scene recently after 4 years in the Fort Dix re-education camp. So take it away Tyler!
Thank you Don, First Ladies and Gentleman I would like to say that this is such a great country. Our population is down to 50 million now with the great matyrdom of the men, women, and children who have valiantly fought the Islamo-fascists in 45 countries around the globe and 6 states of our great Union.
It is a honour to have known some of them personally and to even have seen their deaths on live television. We on the home fronts are not detered and are ready to get down to a 40 million total population, when the great leader calls for it.
Now over in the stock market how can anyone argue with DOW 10^18. Certainly we have seen some minor churn here, but does anyone doubt that Apple won't come through with a new version of E-Orchard and possible a product like V-Resort. I can't wait to check my lottery dividend envelope to see if I get a free ticket to the latest V-sperience.
Now on a personal note some of you may remember me when I was the so-called leader of that Alt. News Page Zero Hedge, well that was an eye opening experience for me, just like I know everybody has rough spots in their rosy lives. But three weeks at Camp Snoopy, and then 4 years at Fort Dix playing cards with Martin Armstrong between E-shock treatments and memory implants took care of that negative attitude. Its ok to get down on America and her greatness, but after awhile you figure out the negativity is in you and that in reality America is always great and will always be great. So if your grief does go on for longer than 6 months call Camp Snoopy, they are the people who care.
/every color of the rainbow is madness./
Surprise!
BP oil spill caused by 'negligence or misconduct', says drilling partnerAnadarko says BP should foot entire bill for cleanup as new estimates show spill could last for up to four more years
http://www.guardian.co.uk/environment/2010/jun/19/gulf-oil-spill-anarkdo...
Oh how quaint. Goldman Sachs thinks people care about them analyzing anything.
they're still putting on the dog and pony show as if they are legit analysts or something, but the world is getting wise to them
http://www.youtube.com/watch?v=mVgY33vGP0g
"Surprise" is in the eye of the beholder...dui bu dui?
Bring back the blog roll. I find that I visit you guys less after I read your stuff. I now go to Mish's site to click blog links. Think about it. Less visits = less page views = less ad clicks - hello!
Blogroll has been removed temporarily until roll out of new server infrastructure/website is completed. It will be returned shortly
...new server infrastructure/website...
Thank you, Jesus!
Its quite awesome If i may say so. I mean the new site that is.
In other words, it is about time economists start factoring in a double dip instead of continuing to live in the land of unicorns, endless stimulus and pixie dust
Tyler, didn't ya know it's Goldman Sachs of pixie dust?
Let no one question our goodness .. in the chief pursuits of our age. The tapping of; Keys, Kegs, Goldman Sachs of pixie dust and other willing members of human society. (Yea John Willmot for the inspiration)
I heard Goldman was shorting pixie dust. Anyone know how to trade this stuff? I smell a bull market.
DOW chart update :
http://stockmarket618.wordpress.com
http://www.zerohedge.com/forum/latest-market-outlook-1
http://www.bloomberg.com/news/2010-06-17/eu-leaders-to-push-global-taxes-on-banks-financial-transactions-at-g-20.html