Picking up the right file

The Institutional Investor advises its readers:

Going public may just be the most important -- and nerve racking -- decision any company will make.  Managing and pricing an IPO is tricky, so picking the right underwriter is crucial.  Bankers often boast of their league table prowess to win mandates, but quantity does not necessarily mean quality.

By quantity, they meant the amount of underwriting fees (revenues) earned; and by quality, the average stock performance of the newly-public companies, as of Feb 16, 2007.

Ten banks were compared on the two Qs using this chart, which is best described as the "file folder chart".

Amusingly, its creator sized the height of each file according to the quality metric, which is the return % listed at the top right corner of each file.  The files were sorted by decreasing quality.  Since each file is a parallelogram, its area is proportional to quality.

However, the files overlap, preventing us from comparing the areas of the files.  Besides, the point made in the article about the importance of both Qs is lost since this chart stressed quality over quantity.  Quantity showed up as a low dot on the tallest file and a high dot on the shortest file.

The junkart version restores the balance.  The blue lines highlighted several banks that scored high on one metric but low on the other.  The construct is a profile chart, with only two variables.

Curious readers may wonder if there were only 10 banks in the IPO underwriting market.  Far from it.  The chart designer introduced a selection bias because banks were included based on Quantity, and then Quality was rated.  This meant there is possibly a boutique firm with small revenues but higher quality than any of the 10 in the plot.

Furthermore, much useful information is missing, including the dispersion of returns, the number of deals, etc.

Reference: "Grading the IPO Underwriters", Institutional Investor, March 2007.

Comments

[wcw]

..of course, equally puzzling is the assumption that high stock performance subsequent to IPO need equate to "quality". All Citi has to do to equal Goldman, after all, is take the same paper it's issuing at $1.00 and price it at $0.85. As the seller, I might well prefer the whole dollar to the Goldman mystique and quality syndicate support.

[meep]

Just for info: what do you use to create your charts?

[Kaiser]

Meep: Most of the charts were created first in Excel, then made pretty in Powerpoint. It does take a lot of time because the default settings are never right. Some of the charts are created using the software R, which is just a great piece of software for doing statistics.

[derek]

I've heard so much lately about people who use Adobe Illustrator to pretty up their graphics that I went looking for a cheap new or secondhand copy. I changed my mind quickly when I saw what the prices were for the Windows version.

So it would be great if one of these days you devoted an entry to sharing some of the secrets of making junk art charts in Powerpoint, or dropped a few tips in when writing up a regular recycling job.

[jussijoo]

Oh my... you're using PowerPoint..!? I'm sorry but from my graphic designer point of view it's purely astonishing that anyone really wants to use that piece of software for drawing purposes.

My first choice is Illustrator, but if you want a more affordable option, why don't you download a copy of excellent Inkscape? It is open source and quite pleasant to use.

[derek]

jusijoo, when you're moving vector graphics from an Excel chart into Inkscape, what do you do to achieve that?

[david]

To see the relationship between "quality" and "quantity" I'd prefer an XY plot. Returns (quality) on Y axis. Actual quantity on X axis rather than merely Rank - probably a log scale. Were the revenues included in the article?
Of course label each data point with underwriter name.
This would show if there is a general trend for quality vs quantity and any exceptions.

[Kaiser]

David: Absolutely, with two numeric metrics, a scatter plot would be a nice alternative. With ranks, I prefer a profile chart.

Derek/Jussijoo: I'll check out those alternatives when I have time. And I'll definitely do a write-up on the charting process.

[derek]

I think the previous commenter was being unrealistically optimistic about importing Excel into Inkscape. It's notoriously impossible without going through e.g. Illustrator, which is no good to anybody when you consider that not having to purchase Illustrator was supposed to be the whole point of using Inkscape. It may be the bees knees for arty types, but it lacks the tools needed for use by most people, unless you can also acquire and learn how to use an entire alternative toolbox chosen for its ability to export into the only formats Inkscape is capable of importing from.

And as soon as you decide to go down that route, Inkscape stops being the logical choice of drawing package anyway. You might as well choose the drawing package that goes with with the stats packages that you've chosen to use.

I'm still interested to learn about your use of Powerpoint for that reason. It may be rubbish, but it comes with Excel, and is what is to hand for those of us used to Excel as a charting tool.

[Cejvik]

Save as ODG and open in OpenOffice Draw. Import excel table and close with save, than open again in Inkscape... Work it?

[Michael Peter]

I love that quote...:

"Going public may just be the most important -- and nerve racking -- decision any company will make. "

Thank you very much :-)

[inkGrapher]

I've plotted a fair bit of data from a spreadsheet into Inkscape. You do have to create your own axes, but when it comes to plotting the data, I take my XY scatter data in two columns, and have a third column as the line segment delimiter 'L'. In inkscape, make a dummy line, then open up the XML editor with Cntrl-Shift-X (I think). Go to the line's point data, simply cut'n'paste the cells from the spreadsheet into the data field.