Previous month:
November 2013
Next month:
January 2014

December 2013

Charles Vest, 72, former MIT President

2010CharlesMVestHeadshot5I was profoundly saddened to hear about Charles Vest's passing of pancreatic cancer a few weeks ago. He was MIT President while I was a doctoral student there and while I didn't have too many interactions with him (we might have attended one or two of the same meetings when I was part of the Graduate Student Council's Executive Committee, but that's about it), I remember chatting with him and his wife at a Christmas reception at their house - it must have been 2002 or 2003.

The obituaries that have been written about Vest emphasize both his fundamental decency and personal warmth as a human being as well as his visionary thinking for MIT in particular and higher education in general. So many years later I remember the same thing - fundamental decency, personal warmth, visionary thinking. Also, a genuine interest in the students who had shown up on his doorstep and a very healthy attitude toward the power inherent to his position - no small feat for someone who led the #1 research university in the world for a total of fourteen years, and later went on to chair the National Academy of Engineering. I guess what appeals to me the most is that, like me, he didn't have an Ivy League pedigree before graduate school but had a family who believed in higher education. He ultimately made a profound difference to the world of research and academia - his ascent from Morgantown, West Virginia boy to MIT President epitomized the American dream. (The path of MIT's current President, Rafael Reif, offers yet another version of that same American dream - that is what I love about MIT.)

Charles Vest will be remembered for a staggering range of accomplishments while he led the Institute. These accomplishments are described in comprehensive fashion in a MIT news release and include his overseeing the fourfold increase in endowment from $1.4 billion to $5.1 billion, a pioneering foray into online education through MIT OpenCourseWare (the predecessor of all MOOCs), international engagement through large-scale ventures such as the MIT-Singapore Alliance, and a string of new buildings, from Simmons Hall to the Stata Center. He is also known for his thoughtful handling of a 1998 report detailing gender-equity problems in the School of Science.

The Boston Globe and the New York Times have obituary notices here and here, respectively.

Vest was also a member of the USA Science and Engineering Festival advisory board, and you can watch him be interviewed for the festival below. (Video uploaded to Youtube by USA Science and Engineering Festival.)

Here is a brief retrospective of his life up to 2012 when he received the 2012 George Brown Prize - one of the many, many prizes he received throughout his career. (Video uploaded to Youtube by internationalscience.)

Finally, below is a longer (45 min) video, uploaded by DistinctiveVoicesBC, where Vest discusses US competitiveness in the 21st century.

I'll give the last word (as quoted by MIT News) to Larry Bacow, who served as Chancellor under Vest before he became Tufts president: "Chuck Vest was, above all, an extraordinary human being. Not only was he perhaps the most respected figure in higher education, he was a man of extraordinary decency, integrity and grace."

"The Launch Pad: Inside Y Combinator"

YCombinatorIf you care about startups, entrepreneurship and innovation, you have to read "The Launch Pad: Inside the Y Combinator" by Randall Stross. The book follows (some of) the startups selected in the Summer 2011 batch of Y Combinator, Paul Graham's Silicon-Valley-based startup accelerator. (And if you care about startups, entrepreneurship and innovation, you don't need me to tell you who Paul Graham is. But if you have arrived at this page because of some other post I wrote, say in healthcare, you might want to know that Graham is an English computer programmer and venture capitalist who co-founded Viaweb, later sold to Yahoo! Y Combinator is said to be the best, or the second best if you ask its main competitor TechStars, startup accelerator in the country. Graham is also an essayist, majored in philosophy at Cornell, studied painting after college and got a doctorate in applied sciences from Harvard, concentrating in computer science.)

What I loved about the book is that it is strongly story-driven with compelling characters - Stross explains well what the various startups are about, but he doesn't get bogged down in technical details (although he could have, given the strong focus of Y Combinator on "hackers"). Instead, he is more interested in showing the dynamics between teams or with Y Combinator's partners, and the reader gets as close as a book can bring to experiencing the summer leading to Demo Day, and to benefitting from Graham's insights. In particular, I felt the book would be particularly relevant to would-be entrepreneurs who not only want a taste of what to expect in startup accelerators, which have been sprouting all over the nation, but also are curious about Graham's thought process, because in that business what matters most for the mentors is to be asking the right questions.

Great read, highly recommended.


I was thrilled to hear last week that Sketchfab, which aims at becoming YouTube for 3D design ("the world is in 3D. the web is in 2D. we bring 3D to the web," with obvious, enormous implications for retailers and anyone who is trying to sell a product online or display a 3D online portfolio, among others), had raised $2 million in its latest round of funding. This brings its total funding to $2.62 million. I like the company not only because it offers a unique service that could change the way objects are displayed on the Internet, but also because it was co-founded by the husband of my visiting doctoral student, both French expats like me.

In only 1 1/2 year of operations (it all started in May 2012 in Paris), Sketchfab has won several business plan competitions and was selected to be part of the Spring 2013 batch of the TechStars NYC startup accelerator program. It is also an alumnus of the Le Camping and WebFWD accelerators. Because the product is web-based, it requires no special plugin. It now has almost 100,000 users and is said to generate more than 3 million page views a month.

You can watch Alban Denoyel - Victoire's husband and Sketchfab's CEO - discuss his company in the YouTube video below, which also displays some stunning examples of what Sketchfab can do. (The two other founders are Cedric Pinson, Sketchfab's CTO and Pierre-Antoine Passet, Sketchfab's CPO [Chief Product Officer].)

You can also read his blog post to Sketchfab users - not only videogame designers but architects, engineers, archeologists and artists - after the company raised its latest new round of funding. (It even has a 3D picture of an Amazon Prime Air drone, uploaded to Sketchfab by user Mestaty.) In Denoyel's words: "We are now a vibrant community of fantastic talents from widespread industries. The models you uploaded have travelled the web, showing it was possible to bring 3D content to a new audience, and establish it as a legitimate media format next to photos and videos."

Watch more Sketchfab videos here. Follow Sketchfab on Twitter at @Sketchfab. Better yet, use the product.

Affordable Healthcare

Cover-in-this-issue-1113One quick mention of the November issue of the Harvard Business Review, since the December issue is already out - the "Globe" feature is on practices at innovative Indian hospitals to deliver world-class health care affordably. What struck me most in this article is that, although the authors - Vijay Govindarajan and Ravi Ramamurti, respectively of Dartmouth and Northeastern University - begin with a modest "India might be the last place on earth where you'd expect to find health care innovation," the hub-and-spoke model it describes is eerily similar to the system strategy guru Michael Porter and his co-author Thomas Lee advocate as one of their six recommendations to fix U.S. healthcare in the previous (October) HBR issue. (I wrote about Porter's and Lee's proposal here.)

In a hub-and-spoke configuration, high-quality talent and specialized equipment are concentrated in urban hubs, while spoke facilities are "arrayed around the hubs to reach underserved patients in far-flung towns and villages." The idea is for the spokes to serve not as mini-hubs but as gateways providing diagnosis, routing treatment and follow-up care. Porter and Lee also advocate a better use of resources by addressing equipment duplication. The authors of the present article, however, do point out significant cultural differences between the U.S. and Indian models that might make the Indian model difficult to implement. For instance, "even when Western hospitals consolidate, their aim is to gain market power rather than to lower costs."

A key element supporting the hub-and-spoke is the increasing use of telemedicine. According to the authors, the concentration of the difficult cases in the hubs has the following effects:

  • Attracting and retaining doctors seeking to improve their skills rapidly.
  • Developing and continually updating treatment protocols that reduce errors.
  • Creating specialists in relativel rare subspecialties of medicine.
  • Promoting innovation that suits local conditions.

Furthermore, general physicians are encouraged to become specialists (and specialists to become super-specialists). The hospitals make significant use of support staff, from paramedics to nurses, to help make physicians more effective. They have also implemented new procedures "to decrease the amount of time it takes to move one patient out of the operating theater and bring in the next one."

The authors also make the fascinating point that in the U.S. (operating in a fee-for-service model), the goal is to maximize the number of procedures conducted, while the goal in India is to maximize the number of patients treated. The Indian hospitals are also frugal in their operations, for instance preferring in some cases to operate diagnostic equipment on a pay-per-use basis rather than owning it outright. This all contributes to an Indian healthcare system where the specific hospitals considered (most definitely best-in-class types) have quality measures -- e.g., survival rates for breast cancer -- similar to the measures in their American counterparts. They also argue that, while physicians' salaries are often cited in the U.S. as the root of many healthcare problems, lower wages in India do not contribute to lower costs as much as one would think. 

The HBR article serves more broadly as a summary of a research project spearheaded by Govindarajan and Ramamurti, which aims at better understanding "how some Indian hospitals are able to provide world-class health care at ultralow cost." Excellent read.

The “Ship to Store” and “Ship to Box” business models

A number of retailers have quietly been implementing an innovative shipping model for their online customers: the free “Ship to Store” option, which is slowly becoming as ubiquitous at the checkout of the retailers’ websites as the more customary “next-day shipping” or “two-day shipping” options. Household names like BestBuy, LL Bean, RadioShack, Wal-Mart all currently offer a “Shop to Store” option, under this or other names such as “Pick Up Today”. The idea appears to be to get more online customers in the stores, so they can make additional purchases when they see the products. This rationale has been explained in multiple news articles, for instance here, here and here. Blending the segments between online and brick-and-mortar customers can also benefit customers: for instance, Nordstrom combined its online and store inventories so that “if the online stockroom is out of a jacket, a store that has it can ship it to the Web customer.”

I became interested in these new shipping models only very recently, when one of my doctoral students got a position as Research Scientist for the Locker team at Amazon in Seattle, WA, starting in the spring. Mind you, I had only noticed the Locker option on the Amazon website weeks before, at the same time as I observed the threshold for free shipping to the customer’s home had increased to $35. Back then I had asked myself what exactly Amazon Locker was supposed to be, but I hadn’t investigated further – I had just selected my delivery option and pressed the “Place Order” button.

Now that I’m going to have a student working for that line of business, I am obviously more interested in the Amazon Locker concept. The purpose, apparently, is for Amazon to take back the “last mile” part of the distribution process. Customers select a locker near their home where they will pick up their item when it arrives, and voila! (An advantage for customers is that the package is secure until they are able to pick it up, in contrast with packages left on doorsteps. UPS has also implemented “Ship to Store” options for that reason.) This obviously offers economy-of-scale opportunities since the shipping of various items remains “bundled” all the way to the end of the shipping chain at the locker site, although the geographic areas where daily postal service is the most in danger (given USPS’s continuing financial troubles) are the rural areas that probably don’t have the demand to begin with to justify a locker. Amazon started rolling out lockers in the Seattle area in 2011, as well as New York City and the Greater DC area. The program expanded in 2012 to Northern California.

This could open remarkable possibilities in the way companies think about shipping items to customers. If I can have faster delivery from Amazon to, say, my neighborhood supermarket than to my home, I would definitely consider shipping to a locker. In contrast with retailers or UPS stores, Amazon does need to enter partnerships with physical stores to put the lockers, so there are also fascinating questions regarding the incentives that should be offered. Participating stores currently receive a small monthly fee, akin to rent. (A possible advantage mentioned early on was that the lockers could help the retailers that “host” them by increasing foot traffic; however, Staples and RadioShack had the lockers removed after a test period.)

While Amazon Locker seems to have mostly initiated partnerships with large retailers such as Staples and stores operating 24-hour-a-day such as 7-Eleven, it’d seem natural to investigate also shipping to the post office (as mentioned earlier, USPS could use the rent money) and installing lockers on high-traffic parts of university campuses/towns. (In fact, students in the San Francisco area have asked for and obtained lockers in locations of a popular coffee chain. Amazon maintains a webpage where customers can submit such suggestions.) Needless to say, the scope of the revenue management research questions that arise is also enormous, starting with demand forecasting and locker location, and of course including the proper amount of rent. (As mentioned in this article, the lockers use a lot of electricity – so that customers will able to unlock their lockers by entering the access code they’ve been given online.) It should make the work of a Research Scientist highly valuable.

Amazon isn’t giving up on the postal office, though – it recently announced that the postal service would deliver on Sundays during the holiday period for Amazon Prime customers in some areas.

MIT Technology Review's "Tale of Two Drugs"

TechReviewNov13The November issue of MIT Technology Review investigates why drugs are so expensive in its cover article. (It also has an excellent business report on healthcare costs.) The two drugs of the article's title are Kalydeco, the first drug to treat the underlying cause of cystic fibrosis, which was approved by the FDA after only 3 months of review, and was priced at $294,000 a year. (Its price later increased to $307,000.) The company did pledge "to provide it free to any patient in the United States who is uninsured or whose insurance won't cover it", but "insurers and governments readily paid the cost" because "it offers life-savings health benefits and there is no other treatment."

In contrast, Zaltrap, which was approved to treat colorectal cancer, "worked no better in clinical trials than Roche's cancer drug Avastin, which itself adds only 1.4 months to life expectancy for patients with advanced colorectal cancer." Yet "Sanofi priced Zaltrap at $11,000 a month, or twice Avastin's price." Following unexpected resistance, it agreed to provide rebates for 50% of the price, which is not quite the same thing as dropping the price by 50%.

The author, Barry Werth, makes an important point when he writes: "The primary customers in the United States are not patients or even individual physicians, although physicians can drive demand for a drug; rather, the customers are the government (through Medicare and Medicaid) and private insurance companies." The prices we see are a result of insurers and governments picking up the tab rather than individuals: increasing the price doesn't reduce how much the drugs are used, so pharmaceutical companies try to charge as much as they can: what they think the market will bear. (Werth knows Vertex well: he wrote the famous book The Billion Dollar Molecule: One Company's Quest for the Perfect Drug about it in 1995.)

Kalydeco's price

But even Kalydeco's price is enormously high, notwithstanding its life-saving characteristics, and Werth discusses at length the concerns and protests of a group of scientists led by David Orenstein, co-director of the Palumbo Cystic Fibrosis Center and chair of the ethics committee at Children's Hospital of Pittsburgh. As a revenue management expert, what struck me in that conversation is that the physicians really wanted Vertex Pharmaceuticals, which manufactures Kalydeco, to use cost-plus pricing in setting the price (i.e., you compute the cost and add a markup because you do want to reward your investors, but since there is no other treatment you don't want to make a fortune from sick people who absolutely need it) instead of revenue (or profit) maximization. This being a publicly traded company, this argument is unlikely to gain traction with Vertex, although there is something fundamentally worrisome about shareholders who hope (do they?) that a pharmaceutical company will price its life-saving drug as high as possible so that they can reap a cushy reward from their investment.

As a mitigating factor for Vertex and Kalydeco (and something I wish the author had explained), drug prices are not only set to recoup past development costs and other drugs' failures. They are set to fund current operations: current development costs, current trials, with many promising compounds bound to fail after years in the pipeline. It also happens that Kalydeco is only designed to treat about 4 percent of the cystic fibrosis patients worldwide or 2,400 patients altogether. (It fixes what is called a gating mutation while most patients have what is called a folding mutation of a critical protein.) I don't know how much of the $307,000 tag price helps fund current operations at Vertex and how much really goes to enrich shareholders, but since it treats a chronic condition, they are looking at a potential income of $720 million a year until the drug goes off-patent, not taking into account price increases. (It is said to be valued now at $311,000.) Whether you view this as high or low depends on what they are doing with the money and what is left over - their net income.

Vertex's financials

Now, Vertex is not a newcomer in the pharmaceutical field. It was in fact created in 1989 and now counts about 1,800 employees including 1,200 in the Boston area. It worked for 13 years with the Cystic Fibrosis Foundation to develop the drug. If you look at its financial numbers on Yahoo! Finance, you'll notice that all its performance ratios are negative. It has a negative operating cash flow. It has negative EBITDA (earnings before interest, taxes, depreciation and amortization). Perhaps this has something to do with the fact that Vertex is building a $800m new campus on the South Boston waterfront. Morningstar gives it four stars but also an F in profitability (and A in growth), and a credit rating of BBB. It was recently announced it is laying off 15% of its workforce due to disappointing sales of its drug Incivek. So the question (left unsaid in the article) also becomes: if a pharmaceutical company hits it big, especially in an area where there is no alternative (and so its discovery is clearly benefiting the public good), how much should we as a society tolerate that it milks its discovery for everything that it is worth, given all the failures and disappointing results that it has had along the way and is still facing in other product lines? (Also worth reading: the JAMA editorial penned by David Orenstein and his colleagues.)

It's worth pointing out that cystic fibrosis patients with the gating mutation can be correctly identified by diagnosis tests, so Kalydeco won't be given to people who won't benefit from it - only to people who will. We used to think that technology would help decrease drug costs by better identifying the subgroups of patients who would most benefit from this or that drug, but the journalist takes a far more pessimistic view of the future, "a likely future in which extremely costly drugs are common."

Going forward

A danger, it seems to me, is that one of these days a good PR professional working for health insurers will put together a campaign where we see sick children being denied their medicine because the greedy pharmaceutical companies are eating up all the insurers' money, with their enormous prices for orphan drugs that go to only a handful of patients - and those are chronically ill to boot. Viewers will relate to the children in the ad and their common condition, and mayhem will ensue. Or put a cystic fibrosis patient on television saying "my $300,000 a year drug goes toward their $800 million brand new building on the waterfront!" (And the PR person will put an executive in the ad who laughs and says: "Heck no, that's what last year's $800 million we made in revenue from people like you paid for! Our palace is paid now! Hahaha!") Then the drug companies risk becoming the Enrons of our generation.

Werth concludes: "Ultimately, what matters in pricing a drug is its value. Vertex succeeded in pricing Kalydeco because the medicine really works, because the company's scientists knew the exact genetic profile of the people who would benefit, because they were able to show definitive clinical results in well-designed trials, and because the company ensured that the right patients got the drug and that access was no issue."