The University of Washington’s book full of (bullshit) metrics

Xconomy editor Benjamin Romano interviewed University of Washington president Michael Young with regard to UW’s technology commercialization program. The conversation is enlightening and worth the read.

Mr. Young is adamant that UW has started all the companies he claims, and explains how UW goes about squaring the difference between claiming to “start” new companies and licensing UW IP to them. The argument apparently is that until UW licenses something to them, they are not real companies:

If I’m saying 18 started this year, next year, say 18 start again, some of which may actually have been incorporated this year, what we’re looking at is that kind of trajectory. Over time, that smooths out. That is to say, if they started four years ago as a company and actually became a real business this year, four years from now there will be a company that started this year and became a real business four years from now, and I’ll count it four years from now, I won’t count it this year.

The use of the first person here is fascinating, as if Mr. Young personally has counted the startups claimed by the Center for Commercialization and determined when they have become “real businesses.” If we go with the AUTM definition of a startup, it is a company formed expressly to take a license to a university’s IP. According to AUTM’s instructions, the startup is counted when it is formed, not when the license is granted. There is nothing about deciding when it becomes a “real business” and no banking of startups to roll out later to help smooth the numbers four years from now into a trajectory that paints the picture one intends the public to see.

On Mr. Young’s argument, how is Marine Construction Technologies (claimed in FY13) a “real business”? Its web site is a single splash screen. How real is that? What about the company cited for FY13 by C4C that was so new that it did not have a name yet? How real is that? What about Spark Medical, apparently started in 2001, which shares its CEO with Marine Construction Technologies (Chairman) and with Resolute Medical (Chairman), and who *also* is the CEO for a Project Canton. Marver doesn’t list his CEO duties with Spark Medical on his LinkedIn profile, but the W Fund lists him. It’s just that Spark Medical appears to have no web site, no operations, no press releases. Nada. Perhaps it has changed its name. Perhaps it is just a shell company. How is it a “real business”? Or how about Universal Cells, created in Jan 2013 (FY13). Its web site is owned by C4C. The company address is a house in Seattle. How is that real? What about PET/X (Jan 2012)? The company web site is down today. PET/X has UW support via a Gap Fund internal grant (that goes to faculty, not to companies). How has that company gotten “real”? Perhaps Mr. Young will count it in four years.

Consider companies that have gotten real. Lodespin Labs, a company started in 2010, was awarded an STTR grant in 2011. If the company is real enough for the US Government to give it a research grant, why is it not a “real business” to be counted in FY11, say? Lodespin may be an “early stage” company, but it was sure not started in FY14, and surely did not become a “real business” in FY14. Deurion started in March 2011 and got an SBIR grant from the NSF in 2011. That looks “real.” Of course, now the company has all but moved to Baltimore. No regional economic development in Washington state for Deurion. Clearly Deurion was “real” well before FY14. Why is UW claiming that C4C started Deurion in FY14? Shockmetrics started in 2008 by UW’s TechTransfer office, before C4C existed–perhaps UW counted Shockmetrics then. C4C advertised Shockmetrics in 2011 at an “Innovation Showcase”:

shockmetrics

Seems Shockmetrics was real enough for UW in February 2011. So was SNUPI, which C4C claimed as a startup in FY13. Now, however, Shockmetrics has no web site, its business address is a house, and its founder speaks of it at his web site in the past tense. How real is Shockmetrics in FY14? Not as real as in FY11, but apparently real enough to pass for “real” by Mr. Young’s standards.

Of course Taggpic was started in June 2012 at Cornell and was acquired by Google, apparently early in 2014. I guess that’s real–but not for UW. Ennaid Therapeutics also was started on the east coast in September 2012. Its offices are in Atlanta and New York. Neither appears to have started to license UW technology, though UW apparently did license some technology to it, according to this website. Both were “real” companies well before FY14. Neither appears to be a “UW startup” by any stretched way of counting.

I could go on. On a first pass, 8 of 18 companies counted by Mr. Young have no reasonable standing to being counted as coming to be “real businesses” in FY14. Of the 10 remaining companies, there are plenty of issues as well. But even with 10, counted willy-nilly based on a whim-based notion of what is real business, means that C4C has not doubled its startup count at all. UW startups are just what they were for the three years before Linden Rhoads dismantled UW TechTrans and created C4C. It’s just that C4C has spent millions more per year than TechTrans ever did to “create” those ten companies. Mr. Young again:

But what seems to me misleading, and I’ll tell you why I don’t count it this year as opposed to four years from now: A lot of companies incorporate and set themselves up and then never do anything. You can look at some universities where you’ve got quite a few companies, but they’re really, honestly speaking, still shells. Now maybe they’ll come to a business and maybe they won’t. I’m more interested in counting when there actually is a sense there’s a business there as opposed to a shell. And I don’t mean shell in a negative way—It’s not a scheme behind it. It’s just that often they’ll have set up some entity that won’t really get its technology, get its marketing plan, get the first product and stuff—that’s the point at which I care.

By counting it that way, what we really are doing is trying to actually be conservative, not particularly aggressive in counting.

Who knows, maybe I had 24 companies start this year, but the only ones I really want to count are those that really have begun in some kind of business-oriented way.

At the University of Utah, Mr. Young presided over an operation that claimed 20 startups a year for six years running. There most of them–90% or more–were shell companies. Business address was the tech transfer office. Web site created by university employees. No ops, no employees. No products, no sales, no economic impact. These companies were only good to attract the envy of other universities’ administrators, create bragging rights, and induce the Utah state legislature to fund expansion of facilities at Utah and Utah State, resulting in–get this–next to nothing in the way of economic impact. Legislators, following an audit, described the claims by the University and the USTAR program as coming from a “culture of untruth and lies.”  The Utah tech transfer program has since melted down, its principals have left, and the university is working to rebuild from the mess created on Mr. Young’s watch as the president of the University of Utah.

AUTM has instructions for how to count startups and distinguish startups from small companies. Here is one version [AUTM has removed the document] (for FY2012, the most recent year available on AUTM’s website):

autmstartups

Here is another version of these instructions, for FY2012 again.

autmstartupsedited

According to the fuller instructions, a startup company is to be counted when it is formed, not when it signs a license. It appears there were some issues, prompting AUTM to clarify how to count startups. The critical question is: “Was the company that licensed a technology formed specifically to license and develop the technology being licensed.” If so, the company should be counted for the year it was “new”–not when it eventually signs a licensing deal. With no licensing deal, by the way, it is hard to see how a new company can be a startup–how can a company depend on a university’s technology for its formation and then not get access to that technology? I suppose the technology could be placed by the university into the public domain. How often do you think that happens? If the technology is not owned by the university, then it cannot be “your institution’s technology” can it?

However one wants to construe AUTM’s definition, there is little to support Mr. Young’s counting method. It is deceptive, inconsistent, and not supportable. C4C is simply making up a good story about itself, one that flatters Mr. Young, and he has been willing to turn that story into a signature media campaign about how wonderful the commercialization thing has been going, with the implication that this wonderful thing is doing good in the world. It’s a nice wish, but it has not been coming true.

Here is Mr. Young’s account of why startups are so important. See if you can make sense of it:

But what I’m trying to measure isn’t so much, have we provided some infrastructure for some guys to go out—although I do. I have a book full of metrics. What we brag about in terms of 18 companies is not by any stretch all I know about what’s going on, on the one hand. On the other hand, what the startups do measure—as the licenses do, actually—is something that I think matters, and that’s probably why we define it the way we do and measure it the way we do.

It would be wonderful to ask for that “book full of metrics” and see what it is Mr. Young is working from. Other than that, we find out that startups measure “something that I think matters.” Well, what is it that UW’s startups measure? Technemort, that which will not be spoken?

So Xconomy asks about criticisms. Here’s the question, and Mr. Young’s answer:

X: There are some critics who read the recent pronouncements about startup formation as the UW padding its commercialization resume a bit, at a time when it’s going to be needing to look for more money to support future commercialization activities in the absence of the Hall patent revenue, and also to fund capital improvements in New Ventures and things like the innovation district.

MY: I’m surprised there are cynics and critics. It shocks me.

I’m going to try to be polite about this—that’s kind of a silly criticism. And the reason I think I can say that with some confidence is that I worked really hard on commercialization and trying to get that ecosystem created at the University of Utah, and I’d never even heard of the Hall patents.

So I think to say that we’re doing this to pad our resume to get this is a little bit bullshit, just to be honest about it.

I’ve got a long track record of caring about this for all the reasons I’ve stated. It matters that we make the world a better place. And this is one of a number of vehicles for doing that. My history entirely belies the fact that the expiration of the Hall patents has anything to do with it.

So, God bless ‘em, but it’s just nonsense.

I would like to meet the critics who have suggested that UW is trying to “pad its resume.” That’s not been the issue at Research Enterprise. Here, we have pointed out that UW is faking its metrics. Not merely being fast and loose. We know what Mr. Young did at Utah, and it was years of fakery that induced the state to pour hundreds of millions of dollars into the University of Utah to support innovation that would lead to economic impact. It didn’t happen. It was a culture of untruth and lies. That is what Mr. Young needs to explain. Utah folks may have been good and snookered, but why should Washington folks think it won’t happen here? Mr. Young might check back about that innovation ecosystem in Utah.

Now, let’s talk about “bullshit.” Harry Frankfurt identifies the essence of “bullshit”:

It is just this lack of connection to a concern with the truth–this indifference to how things really are–that I regard as of the essence of bullshit. (33-34)

As far as I can tell, Mr. Young has a regard for his track record, but not for the truth. If he were concerned with the truth, then he would admit that UW did not start 18 companies in FY14. At best 10, ignoring AUTM’s instructions. More accurately, based on what C4C has published, 1. That’s the truth, as best as public records provide it. Mr. Young may have a “book full of metrics,” but that book–I am being sharp here–is bullshit. What’s nonsense is C4C’s reckless spending, false reporting, absence of oversight, and disregard for laws. And all for what? A charade to snooker the state legislature into giving UW millions for “economic development” or “innovation”? To make C4C “too important to fail”?

 

This entry was posted in Bozonet, Metrics, Startups. Bookmark the permalink.

Leave a Reply