Happily, I’ve only used the term “green shoots” one time in the brief history of TGR, and then only sarcastically in the title of this cartoon (which I stand by, as this post should make evident).
The term has always struck me as ridiculous, and not solely because it was first uttered at a time when it was not only false, but utterly misleading. What’s worse is that the manner in which the media has pounced on the phrase has eliminated any shades of meaning, much as our eyes glaze over as reports of “billions of dollars lost” and “hundreds of thousands of jobs eliminated” come out — we have become desensitized by the magnitude of the concept and our overexposure to it (not to mention that no matter how many times we shut our eyes and whisper, it doesn’t seem to materialize).
Ultimately, the term has become synonymous with the “second derivative” argument – things are getting worse, but they are getting worse at a slower rate – green shoots sprouting! And while I don’t at all equate “not-as-bad news” with “good news”, I was happy to let the second derivative camp savor their banner phrase.
Until this morning.
For some reason, today I finally began to think about what “green shoots” really means: it represents the spring, rebirth and growth. It doesn’t stand for a positive second derivative, but for a positive first derivative – something universally aknowledged not to be the case. I find this revelation infuriating: if we don’t have a positive first derivative, representing growth, then how can there be green shoots, which also represent growth?
For those willing to continue reading, I’ll illustrate what I mean with graphs that may confuse more than they educate. Shall we? Let’s shall.
Follow a plant through it’s life cycle: it grows in spring, flourishes in summer, withers in the fall and essentially hibernates in the winter (I don’t know what the proper horticultural term is). Since I want to tie this back to derivatives and such, let’s get some math involved. A simple graph of the flower’s height above the ground might follow a sinusoidal curve and, courtesy of Wolfram Alpha really coming through, look like this:
Here is its first derivative:
And here is its second derivative:
In all these graphs, 0 is winter, 1 is spring, 2 is summer, 3 is fall, and 4 is winter again. Also, a key point is that because this is a graph of height above the ground, green shoots would be observed somewhere between 0 and 1, as the plant first emerges from the soil.
Now we need to figure out where we are in this hypothetical plant lifecycle. We know we have a negative first derivative, which puts us between 2 and 4 (summer and winter). We also have a positive second derivative – for argument’s sake – which limits us to sometime after 3 (fall). So we are in the space between fall and winter; our economic “plant” is withering away, albeit at a slower pace than it was during the first cold snap.
So, IF the plant metaphor holds (and let’s assume it does, for why else would we use the term “green shoots”?) and IF we are seeing the second derivative turn positive (and I’m not ready to aknowledge that, yet, but the green-shootists are) and IF the first derivative remains negative (no doubts there), we have not yet made it to spring. Only as we reach spring does the first derivative turn positive and green shoots emerge. Just to be absolutely clear: there are no green shoots yet.
(You’re right, I could have spared you and written that much earlier, but I wanted to use the graphs.)
You will notice that in the winter, the plant actually retracts back into the ground, but I suppose “brown shoots” or the titular “dead shoots” doesn’t quite capture the spirit of that positive second derivative. I’m sure there must be other plant metaphors, like “winter blossoms” or “the last leaves to fall”, that are more appropriate.
I suggest ”pushing up daisies”.
Silicon Valley Insider presented this as its Chart of the Day today, saying it indicates the success of Microsoft’s “Laptop Hunter” ads:
First of all, it takes some digging to learn what this scale even means, which brings us to a violation of charting rule #1: do not use a misleading axis! The true scale goes from -100 to 100, as some Googling will reveal, so why does the graph go from 0-70? Probably, sadly, for dramatic impact. A zero score means people have as many positive comments as negative; 100 and -100 presumably represent purely positive and purely negative comments, respectively.
Next, consider the volatility of the chart – the standard error of these estimates must be enormous. Apple’s “downfall” crosses a distance that it recently rose in just one week. Again, this doesn’t make the chart wrong, it just makes it difficult to asses whether Apple’s downward move is an increase in negative comments or a decrease in positive comments following an abnormal burst of them in mid-March.
The best evidence for some sort of regime change is that the two firms are closely correlated for the first half of the chart, and negative correlated for the second half – though, without proper analytics, it’s hard to see how “real” the effect is. But the early beta moves and later alpha moves suggest that – to the extent the chart is “real” – one firm or the other experienced some sort of idiosyncartic event near the middle of the chart.
But what really bugs me is the annoations the SVI added to the chart (not part of the original presentation). The spike for Apple isn’t the new iMac (which was underwhelming released with zero fanfare), it’s the preview of the iPhone 3.0 software!
Seven years ago, the Mets were among the first teams in MLB to adopt a tiered pricing system in which it costs more to see a game against a good opponent than a bad one. At the time, it was the most sophisticated such plan in baseball. Others included simple methods like charging more for weekend games or prime summer games.
The act was a major step toward recognizing that while the supply of tickets was more or less constant for every game, demand could vary wildly. Nonetheless, ticket prices were still set for every game before the season started, and were locked in no matter how attrative (or otherwise) the game ultimately appeared. Rain, cold, chances to see historial achievements, unexpected opponent records (good or bad) – all of these affect demand in ways which can not be anticipated in March.
Now, the Giants have announced a new initiative which will adjust ticket prices up until the first pitch. A software program will estimate demand based on dynamic variables and adjust ticket prices according. The factors include both teams’ records, the stats of individual players (including the starting pitchers), the weather, the number of seats left to sell, proximity to gametime, promotional nights, and other similar metrics. Fans showing up on a rainy weekday night just before the game might be able to snag tickets for $5 that were $30 the week before.
It is likely that the model is not nearly as complex as one might think – in fact I imagine this is a perfect example of a time when a simple model can account for a surprising amount of variance. Not that this approach is revolutionary outside MLB – airlines and hotels have been doing something similar for years – but I think its an exciting and approrpriate use of data and would not be surprised if next season the metric is rolled out in a more expansive manner.
To think it was just a decade ago that the concept of statisticians on the field was inconceivable (but taken for granted today) – and now we have teams employing econometric methods to estimate demand. Quelle journee!
April advance retail sales were announced much lower than expected, coming in at -0.4% vs the anticipated 0% month-over-month change. Auto sales constitute a large part of retail sales (around 20%, in fact), and so it can be informative to look at the retail number excluding autos to see the real trend in consumer behavior. That figure was -0.5% vs an expected 0.2% gain.
You may recall that not long ago, the seasonally-adjusted auto sales rate for April was announced at a dismal 9.3mm vehicles, following a March rate of 9.9mm (and an awful February, 9.1). This looks like a clear case of auto sales dropping in April, which is unsurprising given the Chrysler/GM turmoil. Even Toyota sales were hurt as consumers shied from the entire industry.
The economists who are surveyed before the sales figures are announced clearly took the poor April data into account, since their median sales number inclusive of autos is 0.2% lower than ex-autos. But the actual data released this morning (-0.5% ex-autos; -0.4% inclusive) implies that auto sales contributed a gain of 0.1% in April!
Maybe we’ll get a revision?
Caveats: advance sales are noisy estimates based on incomplete data; advance sales figures are released by the Census Bureau while monthly auto SAARs are estimated by AutoData Corp.
In response to thoughts that Ty Cowen’s new economics textbook might threaten his income, Greg Mankiw amusingly responds:
It is true that there is always entry into the textbook-writing business, which imperils incumbents like me. But I am not worried. I am one of the economics profession’s leading producers of textbooks, I have an extensive network of dealers (aka professors), and I have friends in high places (Larry Summers, Christy Romer). So doesn’t all this make me precisely the kind of too-big-and-too-interconnected-to-fail plutocrat that, if push comes to shove, will get a government bailout?
If you doubt me, let me point out that my initials are GM.
When I was in school, we read a paper on “damaged goods” – products which a manufacturer has intentionally disabled in order to price discriminate or drive revenue elsewhere. Now, Jorge Garcia’s blog (you may also know him as Hurley), demonstrates the phenomenon in the wild via this accidental economics lesson:
Kudos Smarte Carte
On the new design.
Not only is the newer cart shorter in length, but it also doesn’t have that raised lip at the end to hold luggage in, like the old one did.
Now it’s almost IMPOSSIBLE to balance even two bags on it.
So more people will be forced to get a second cart.
Cha-Ching!
You can check out Jorge’s blog
here.
Caught this on Rortybomb - Barry Eichengreen has penned an excellent piece on the role of models in academia and finance, as well as the growing importance of empiricism (a point with which I particularly empathize). An excerpt:
Maybe so. But amid the pervading sense of gloom and doom, there is at least one reason for hope. The last ten years have seen a quiet revolution in the practice of economics. For years theorists held the intellectual high ground. With their mastery of sophisticated mathematics, they were the high-prestige members of the profession. The methods of empirical economists seeking to analyze real data were rudimentary by comparison. As recently as the 1970s, doing a statistical analysis meant entering data on punch cards, submitting them at the university computing center, going out for dinner and returning some hours later to see if the program had successfully run. (I speak from experience.) The typical empirical analysis in economics utilized a few dozen, or at most a few hundred, observations transcribed by hand. It is not surprising that the theoretically inclined looked down, fondly if a bit condescendingly, on their more empirically oriented colleagues or that the theorists ruled the intellectual roost.
But the IT revolution has altered the lay of the intellectual land. Now every graduate student has a laptop computer with more memory than that decades-old university computing center. And she knows what to do with it. Just like the typical twelve-year-old knows more than her parents about how to download data from the internet, for graduate students in economics, unlike their instructors, importing data from cyberspace is second nature. They can grab data on grocery-store spending generated by the club cards issued by supermarket chains and combine it with information on temperature by zip code to see how the weather affects sales of beer. Their next step, of course, is to download securities prices from Bloomberg and see how blue skies and rain affect the behavior of financial markets. Finding that stock markets are more likely to rise on sunny days is not exactly reassuring for believers in the efficient-markets hypothesis.
The data sets used in empirical economics today are enormous, with observations running into the millions. Some of this work is admittedly self-indulgent, with researchers seeking to top one another in applying the largest data set to the smallest problem. But now it is on the empirical side where the capacity to do high-quality research is expanding most dramatically, be the topic beer sales or asset pricing. And, revealingly, it is now empirically oriented graduate students who are the hot property when top doctoral programs seek to hire new faculty.
Not surprisingly, the best students have responded. The top young economists are, increasingly, empirically oriented. They are concerned not with theoretical flights of fancy but with the facts on the ground. To the extent that their work is rooted concretely in observation of the real world, it is less likely to sway with the latest fad and fashion. Or so one hopes.
The late twentieth century was the heyday of deductive economics. Talented and facile theorists set the intellectual agenda. Their very facility enabled them to build models with virtually any implication, which meant that policy makers could pick and choose at their convenience. Theory turned out to be too malleable, in other words, to provide reliable guidance for policy.
In contrast, the twenty-first century will be the age of inductive economics, when empiricists hold sway and advice is grounded in concrete observation of markets and their inhabitants. Work in economics, including the abstract model building in which theorists engage, will be guided more powerfully by this real-world observation. It is about time.
Rortybomb does have one point with which I disagree, that being that VaR is a statistical (i.e. empirical) rather than theoretical figure. This is true only inasfar as the VaR is a historical VaR calculated from past returns; estimating a probable future VaR (as banks do in their financial disclosures) is a highly theoretical exercise in estimating the copula of portfolio returns – both the marginal distributions of each asset as well as the dependence structure. A minor point but, I think, an important distinction. Anyway, check out the full piece, it really is one of the best essays I’ve read in a while.
The first week of results on iTunes’ tiered pricing are in and they are positive. According to a Billboard report, songs which experienced price hikes of 30% sold only 12.5% fewer units (only 6.9% fewer if you ignore “the expected second-week drop of Black Eyed Peas’ Boom Pow Pow” – a wholly un-justified remark which this statistician is forced to ignore). This means that digital tracks do in fact exhibit price elasticity, a belief I’ve held without verification for many years. In fact, the elasticity appears to be greater than 2 (30/12.5). However, a few caveats:
- The sample is ridiculously small and the data extremely noisy
- Songs in the sample that remained at $1 sold 9.9% more units than the previous week
- Sales of all digital tracks increased 3% week-over-week, and songs in the top 100 increased 1%.
If the control group’s sales really increased 9.9%, then the drop of 12.5% in the higher-price group is actually a drop of 22.4% relative to where they should have been – yielding a much smaller elasticity (though still greater than 1). The true number is likely somewhere in between, since the price hike probably drove customers to purchase the control group (making it not a control group at all!)
Using a number of 6%, which is between the control group’s 9.9% and the population growth of 3%, the elasticity is 1.6. It’s too early to draw conclusions, however.
Slate has an interactive map which illustrates job losses by county throughout the US over the last two years. It’s very sobering to watch the red circles (representing losses) explode in late 2008.
It’s hard to believe FT Alphaville is taking this seriously, but they are: markets and sunspot cycles. Apparantly, as this very convincing graph shows, recessions correspond with the regular sunspot cycle:
As this plainly demonstrates, there is a perfect correlation with sunspots and recessions. Except for that little recession in the 1930′s, but that one doesn’t count, right? And this isn’t the first time that sunspots have been tied to the economic cycle – researchers have found an impact on the price of wheat.
What I see here is an overlay of two cyclical occurances, and a somewhat forced conclusion of causality based on their correlation (have we learned nothing?). While the wheat price study is somewhat more convincing, is it such a stretch to think that maybe wheat prices and recessions are linked, and that the sunspots are a spurious correlation that really have nothing do to with either? That argument can be made with equally sound “analytics” (by which I mean looking at pictures).
