“If you have to ask”: Ten sure-fire ways to lose money on research

Normally collegial discussions took a nasty turn after I suggested that most universities lose money on sponsored research.

Incredulous: “I don’t believe it. My department tacks a 50% surcharge to all my contracts; how can they lose money?”

Defensive: “Here are all the reasons that doing research is a good thing, so what’s your point?“

Defensive with an edge: “Why are you attacking research?“

Let’s be be clear about it:  if it’s your institution’s mission to conduct research, then spending money on research makes perfect sense.  In fact, it would be irresponsible to deliberately starve a critical institutional objective like research.

On the other hand, there are not all that many universities with an explicit research mission.  But there is an accelerating trend among  primarily bachelor’s and master’s universities to become — as I recently saw proclaimed in a paid ad — the next great research university. The university that paid for the ad has absolutely no chance to become the next great research university.  Taxpayers are not asking for it.  Faculty are not interested. Students and parents don’t get it either.

The administration and trustees think it’s a great idea.  Research universities  are wealthy.  Scientific research requires new facilities and more faculty members.  Research attracts better students. Best of all, federal dollars are used to underwrite new and ambitious goals. Goals that would be out of reach as state funding shrinks. As often as not, the desire to mount a major research program is driven by a mistaken belief that sponsored research income can make up for shrinking budgets. It’s a deliberate and unfair confounding of scholarship and sponsored research

If your university is pushing you to write grant proposals to generate operating funds, then alarm bells should be going off.  Scholarship does not require sponsored research. Chasing research grants is a money-losing proposition that can  rob funds from academic programs.  It’s an important part of the mission of a research university, but for almost everyone else, it’s a bad idea.  It’s a little like shopping on Rodeo Drive:  there’s nothing there that you need, and if you have to ask how much it costs, you can’t afford it.

How is it possible to lose money on sponsored research?  After all, professor salaries are already paid for.  The university recovers indirect costs. Graduate and undergraduate students work cheap.

A better question is how can anyone at all can possibly make money on sponsored research. Many companies try, but few succeed.  A company that makes its living chasing government contracts might charge its sponsors at a rate that is 2-3 times actual salaries. Even at those rates, it is a rare contractor that manages to make any money at all.

On the other hand, a typical university strains to charge twice direct labor costs.  Many fail at that, but the underlying cost structure — the real costs — of commercial and academic research organizations are basically identical.  There is a widespread  but absolutely false assumption that underlying academic research costs are lower  because universities have all those smart professors just waiting to charge their time to government contracts. The gap between what universities charge and what sponsors are willing to pay commercial outfits is the difference between making a profit and losing a lot of money. Just like intercollegiate athletics, sponsored research programs tend to lose money by the fistful.

Let me say up front that the data to support this conclusion are not easy to come by.  Accounting is opaque. Sponsors know a lot about what they spend, but relatively little about what their contractors spend.  It is in nobody’s interest to make the whole system transparent.  But my conversations with senior research officers at well-respected research universities, paint a remarkably consistent picture.  With very few exceptions, it takes $2.50 to bring in every dollar of research funding.

Fortunately, the arithmetic is easy to do.  If you know the right questions to ask, you can find out how much sponsored research is costing your institution. Here are ten sure-fire ways to lose money on sponsored research. You do not need all of them to get to a negative 2.5:1 margin.  If you are clever just a couple will get you there.

1. Reduce senior personnel productivity by 50%: university budgets are by and large determined by teaching loads, a measure of productivity. It is common to adjust the teaching loads of research-active faculty. Sometimes normal teaching loads are reduced by 50% or more.  It is, some argue, table stakes, but a reduced teaching load is time donated to sponsored research because funding agencies rarely compensate universities for academic year support.
2. Hire extra help to make up for lost productivity: Courses still have to be offered, so departments hire adjuncts and part-time faculty.
3. Do not build Cost of Sales  into the contract price: The sales cycle for even routine proposals can be  months or years.  Time spent in proposal development converts to revenue at an extraordinarily small rate. In nontechnical fields and the humanities where research support is rare, the likelihood of a winning proposal is essentially zero.
4. Engage in profligate spending to hire promising stars: Hiring packages for highly sought-after faculty members can easily reach many millions of dollars.  A sort of hiring bonus, there is little evidence that this kind of up-front investment is ever justified on financial grounds.
5. Make unsolicited offers to share costs: Explicit cost-sharing requirements were eliminated years ago at most federal agencies.  Nevertheless, grant and contract proposals still offer to pay part of the cost of carrying out a project.
6. Allow sponsors to opt-out of paying the indirect  cost of research: An increasingly common practice is to sponsor a research project with a “gift” to the university.  Gifts are not generally subject to overhead cost recovery, so a university that agrees to such an arrangement has implicitly decided to subsidize legal, management, utility, communication, and other expenses, and
7. Accept the argument that indirect costs are too high: The  meme among federal and industrial sponsors is that indirect costs are gold-plating that must be limited. Rather than believe their own accounting of actual costs of conducting research, they argue that universities, should limit how much they charge back to the sponsor.
8. Build a new laboratory to house a future project: Sponsors argue that it is the university’s responsibility to have competitive facilities.  But that new building is paid for with endowment funds or scarce state building allocations that might have gone toward new classrooms or upgraded teaching labs.
9. Offer to charge what you think the sponsor will pay, not what the research will cost:  Money is so tight at some funding agencies that program managers are told to set a (small) limit on the size of grants and proposals independent of the work that will be actually be required.
10. Defray some of the management costs of the sponsoring agency: It has become so common that it is hardly noticed.  University researchers troop into badly-lit conference rooms to help program officers “make the case” to their management.

The list goes on. It is so easy to turn a sponsored research contract into a long-term commitment to spend money for which there is no conceivable offsetting income stream that institutions routinely chop up the costs and distribute them to dozens of interlocking administrative units.  The explosion in the number of research institutions has all the elements of an economic bubble.

• It is motivated by a gauzy notion that all colleges and universities are entitled to federal research funds..
• It is fed in the early stages by accounting practices that make it easy to subsidize large expenditures.
• It has the cooperation of funding agencies who know that the rate of growth is not sustainable.

Virtually everyone involved in university research knows that the bubble will burst.  A colleague just showed me an email from his program director at a large federal research agency.  It said that — regardless of what he proposed — the agency was going to impose a fixed dollar amount limit on the size of its grants. But in order to win a grant, he had to promise to do more.  His solution: promise to do the impossible in two years instead of three.  Just like the famous Sydney Harris cartoon,  a miracle is required after two years. At least there would be enough money to pay the bills while a new grant proposal was being written.

What’s for Lunch?

Awhile ago, I mentioned India’s plan to create 27,000 new colleges and universities over the next decade.  Well, guess what?  I was wrong.  The number is now 35,600. Here’s what I said a year ago about Education Minister Sibal’s plan to expand India’s capacity in higher education:

What does this have to do with American colleges and universities? Just as low-cost, high value service industries have migrated to India, the higher education market in the US will also start to buy more educational services there as well.

So I was immediately drawn to yesterday’s Business Week article about California’s intention to make a quick lunch of its seed corn by cutting university spending $1.4B and the likely effect that snack will have on job growth and tax revenue.

Particularly striking to me was VC Robert Ackerman’s reaction to the massive and rapid expansion of higher education in Asia:

Right now, if I were the Chinese university system, I’d be running ads showing up on UC websites, recruiting students to universities in Beijing and Shanghai.

Now I am not a big fan of the proposition that value in higher education can be measured in dollars spent–if American institutions made better use of their budgets, then the resulting efficiencies would actually increase capacity–but there is little doubt that wholesale dismantling of universities across the country is a very bad idea.

We are shrinking university capacity at a time when India, China, Singapore and many other countries are  increasing theirs. India  alone will create 600 new research universities. China is increasing its capacity in research universities while the  U.S. has created one new research university this century: UC Merced.  Since Merced is part of the California system, its prospects are dimmer by the moment. Only a handful of new universities of any kind have been created in the U.S. since 1960, a period in which college enrollments have quadrupled.

Why is falling capacity so important? Because the worldwide market is growing, and we are systematically reducing our share of that market when economic competitors are  moving in the opposite direction. I leave it as  a homework exercise to determine what happens when an enterprise loses market share in a growing market.

Buon appetito!

The Internal Start-up

I had a conversation the other day with a senior executive — let’s call him Bob —  of a Fortune 10 company about their “internal start-up” culture. It seems that they are looking for breakthrough product ideas that do not align well with their core business.  The solution seems obvious: let’s create the same kind of  exciting, market-driven environment that you would find in a start-up!

Everything sounded fine for a few minutes.  They thought that the most creative people in the organization needed to have elbow room that would be difficult to achieve in the risk-averse culture of a hundred billion dollar company.  So how did they plan to achieve that?

• Freedom to break some rules:  the start-up can use its own  product roadmaps and sales strategies
• Freedom from process-driven corporate calendars and budgets: the leadership of the start-up is not bound by the revenue and earnings goals of their parent
• Freedom to take risks: they have permission to fail

It didn’t take long for the discussion to go seriously off track.  When I started in with questions about how they were going to actually pull this off, Bob said: “Look, I’m in charge of new technology and platforms and I’m going to be the venture capitalist funding a new product, so that when it succeeds we’ll be able to fold it back into our current business.” I had seen this movie before.  It’s called When Worlds Collide. When I suggested that Bob lives on a different world and would make a terrible venture capitalist, things got a little heated. As I recall it, Bob said, “In your ear!” A surefire way to put a fine point on your argument.

Bob lives on a planet where the scale of his business creates a climate for successful development of new products that can be sold to familiar customers using existing channels and tried-and-true processes.  Above all, in Bob’s world, it is possible to make big bets. The examples are impressive. Everything from HP’s inkjet printing to the Boeing 777. Unfortunately for Bob and his start-up, none of those things matter.  The start-up lives in a world of new markets, which means new customers, new channels and new processes.

Even though Bob has all the talent he needs for market success,  the likelihood of failure is high. The Newton and the Factory of the Future did not fail because  because Apple and GE could not innovate.  They failed in large measure because corporations foster a system of beliefs that is fundamentally incompatible with  taking capabilities to new markets. When I asked Bob  how the start-up employees were going to be recruiteed and rewarded, whether they had a safety net for returning to the company in case of failure, and how many simultaneous bets he was willing to place, the answers were not encouraging.

I immediately did a deep dive into my archives, hoping to find traces of a long-forgotten venture that I helped steer into the ground.  In the late 1990s Bellcore was poised to enter the online services business, hoping to attract newer, smaller customers than the seven  Regional Bell Operating Companies who accounted for most of the company’s revenue.  This was a time when Bellcore’s Applied Research group was generating a blizzard of patents in e-commerce and software, technology that I have talked about before. We were as smart and nimble as any West Coast start-up, and best of all we had the cash to fund a new venture, the talent to staff it, and the power of an existing sales team to go after those new customers. I was asked to lead the new company.  We would be funded just like a VC-backed start-up…

When the dust settled and I reported lessons learned to the Bellcore’s CEO Richard Smith and later to Bob Beyster, CEO of SAIC,  Bellcore’s parent company, the first thing I said was that there had been no structural reason for failure.  A team from McKinsey had already given us the range of possibilities. We could have set up an independent business unit or spun 0ut a company in which we retained minority ownership.  Setting up a new incubator would have required more time than we thought we had, and, in any event,  Applied Research was already in the incubation business. We had chosen to bypass corporate reporting structure and create a company-within-a-company with direct oversight by a CEO who was committed to our success.  It was exactly the Hughes DirecTV model.

There are three reasons that internal start-ups like ours tend to fail.  Bob was not in the mood to listen because he is banking on success, but the topic comes up in every large enterprise, so I thought it might be a good time to repeat the conclusions here:

1. Failure is common: Building new business is a portfolio game in which 90% of the returns come from 15% of the investments.  It is fundamentally unlike product development. A “big bet” strategy only succeeds when there is high degree of confidence in your ability to sort out winners and losers.  In a new market, that just never happens.
2. Market-driven milestones drive success in new ventures.  An internal start-up — even one with strong support at the top — cannot divorce itself from processes that are timed to fit corporate needs.
3. Corporate sponsors of new ventures and VCs have different belief systems.  They are fundamentally incompatible, and without early, explicit steps to stop it, corporate attitudes, practices, and beliefs will overwhelm the fragile culture of the start-up.

I want to spend the next several days elaborating on these ideas.  I hope Bob is reading.

Technology Hype and Investment Mania are Not Always Irrational

It’s funny how the same reading of  history leads to different conclusions. The young investor in the 1840s Punch cartoon above stands in a back alley outside the Capel Court stock exchange asking a purveyor of dubious scrip how to honestly make £10,000 in railways. It is the end of a technology hype cycle in which the modern-day equivalent of $2 trillion was pumped into an investment bubble.  The picture on the right is a desolate and economically insignificant outpost connected by some of the 2,148 miles of railway capacity that entrepreneurs built during the British railway investment mania of the 1830s. The conclusion is that early investors in British railway companies were played for suckers.

The mania probably started with an announcement in the May 1, 1829 edition of the Liverpool Mercury:

“To engineers and iron founders

The directors of the Liverpool and Manchester Railway hereby offer a premium of  £500 (over and above the cost price) for a locomotive engine which shall be a decided improvement on any hitherto constructed, subject to certain Stipulations and Conditions, a copy of which may be had at the Railway Office, or will be forwarded. As may be directed, on application for the same, if by letter or post paid.

HENRY BOOTH Treasurer Railway Office, 25 April 1829

The Liverpool and Manchester Railway was not the first railroad in England, but the competition drew enormous interest.  Contestants used everything from “legacy technology” — horses on treadmills — to lightweight steam engines that could reach up-hill speeds of 24 miles per hour. The legacy technology defeated itself when a horse crashed through a wooden floorboard. It did not hurt that Queen Victoria declared herself “charmed” by the winning steam technology.

Business innovation  — ticketing, first-class seating, and agreements allowing passengers to change carriers mid-trip — was rapid and fueled as much by intense competition as by a chaotic, frenzied stock market in which valuations soared beyond any seeming sense of proportion, causing  John Francis in 1845 to despair: “The more worthless the article the greater the struggle to attain it.” When the market crashed during the week of October 17, 1847 — in no small measure due to to the 1845-6 crop failure and potato famine — and established companies failed, financiers like George Hudson were exposed as swindlers. Thomas Carlyle demanded public hanging.

The collapsing bubble is not the end of the story. Between 1845 and 1855 an additional 9,000 miles of track were constructed.  By 1915 England’s rail capacity was 21,000 miles.  British railways had entered a golden age. The lesson that observers like Carlotta Perez and others draw is that there is a pattern to technological revolutions:

1. Innovation enables technology clusters, some  of which transform the way that business is done.
2. Early successes and intense competition give rise to new companies and an unregulated free-for-all that leads to a crash.
3. Collapse is followed by sustained build-out during which the allure of  glamor is replaced by real value.
4. This leads to a golden age that results in more innovation as lives are structured around the new technology.

This is a Schumpeterian analysis of innovation that is reflected everywhere, but particularly in the economics of the new technologies of the late twentieth century.  The stamp of the the 1840s British railway mania can be seen in Gartner’s technology hype cycle and in nearly every discussion of the 2000 dot-com collapse.  It is an analysis that is a special problem for angel and other early-stage investors because there is no real guide to tell you when the bubble will burst. Unless you are George Hudson, what investor will find the risk acceptable? A rational early investor will steer clear of technologies that radiate this kind of exuberance.

But what really happened to all that investment in the 1830s? I was amazed to see the recent article by my long-time colleague Andrew Odlyzko at the University of Minnesota who analyzes the British railway mania example and concludes that the early investments did quite well:

The standard literature in this area, starting from Juglar, and continuing through Schumpeter to more recent authors, almost uniformly ignores or misrepresents the large investment mania of the 1830s, whose nature does not fit the stereotypical pattern.

Andrew enjoys taking contrary — often cranky but always well-thought out–  positions on conventional wisdom, so I approached his article with cautious interest.  After all, I thought I knew a little about the railway mania episode.  I had used it myself to illustrate innovation cycles. Like most people, I had focused on the disaster of the 1840’s, so I was drawn immediately into Odlyzko’s argument that during the mania of the 1830’s,  “railways built during this period were viewed as triumphant successes in the end.”:

After the speculative excitement died down, there was a period of about half a dozen years during which investors kept pumping money into railway construction. This was done in the face of adverse, occasionally very adverse, monetary conditions, wide public skepticism, and a market that was consistently telling them through the years that they were wrong.

In other words, the end result of the wildly speculative exuberance of the  1830s was the “creation of a productive transportation system that had a deep and positive effect on the economy.” Investors saw great returns. A shareholder in London and South Western Railway (LSWR) who in 1834 paid a £2 deposit on a share worth £50 and who paid all subsequent calls (totaling £95.5) would have watched the investment grow to 2.31 shares valued at  £200 by mid-1844 and would have received in 1843 alone £4.62 in dividends — a 9.68% annual return.  This defied the more rational demand and cost forecasts:

at the start of the period…in June 1835, such investor would have paid £10, and seen the market value it at £5.5. In fact, over most of the next two and a half years, the market was telling this investor that the LSWR venture was a mistake, as prices were mostly below the paid-up values.

Andrew Odlyzko is a seasoned mathematician who knows better than try to prove a general principle by example.  He says as much in his paper. On the other hand, railway mania has been used for years as an illustration of an innovation cycle, and  Odlyzko has a very different reading of history. The conclusion that is usually drawn from the Railway Mania may lead markets and investors astray because it seriously misrepresents actual patterns. The whole point of a cycle — hype, innovation, or investment mania — is that it can be used as a risk-averse template for rejecting sales pitches that start with “This time is different“.  But that does not mean that it is never different.

Loose Cannons, Volume 1

This is my all-time favorite Dilbert cartoon. Anyone who has ever worked in a large corporation like Hewlett-Packard understands immediately what’s going on here.  I always used it in CTO coffee talks when I wanted to show our engineers that I was really one of them — that I  wasn’t from another world (although I  suspected that many of them were already convinced that I was the pointy-haired boss and some thought I was Blob).  After a few hours, like clockwork, the email would start pouring into my inbox.  The subject line was always something like: “From a Loose Cannon.”

Some of the messages were very strange and a few (like the ones talking about contacting aliens from space) were downright disturbing, but most of them were respectful notes to let me know of  legitimate ideas that hadn’t made it through internal management gates.  I knew the engineering managers well.  They were smart and careful and for the most part they were very successful.  I didn’t want to second-guess their investment decisions, but I started wondering whether another sort of investment analysis would give a different answer, because these were obviously colliding worlds.

I was not popular with some of HP’s general managers because I had invented a new sort of escalation path for engineers, inviting ideas that had already been turned down at some point in the management chain.  I created a Technology Council consisting of the CTO’s of each of the major business units, the Director and Chief Scientist from HP Labs and some  HP Fellows to help with technology strategy and road-mapping, so it made a great deal of sense to use this team to take one more look at some of the Loose Cannon Ideas.

One of the Loose Cannons proposed using HP’s Jornada Pocket PC “to control my TV and VCR or other IR devices – that way you could store stuff in there and program those things simply and easily.” Another L-C wanted to create a document management system for the “growing home genealogist market”.

The company already had a rich history of encouraging risk-taking by its technical staff, but at HP business objectives were never far from sight.  There was a 60-year history of combining risk with rational investment.  It was a strategy that worked well.  It was lightweight, and I think that’s why cool new products and sometimes whole new product categories continued to flow out of R&D activities.  I am not only talking about the research labs. At that time there were over  12,000 engineers, many of whom had advanced degrees and were rewarded for patents, publications and other creative work; there was incredible bench strength. I will have more to say in later posts about how this process of identifying and nurturing creative ideas was carried out, but today I want to concentrate on the very specific calculation that virtually all R&D managers in the company learned.  I think that the legendary Joel Birnbaum was responsible for it, but my friend Stan Williams, who for many years now has guided HP’s nanotechnology and quantum computing research nailed the analysis in a dramatic way[1]:

…Why don’t we put together a program to become the world’s best center in quantum computation?

The answer is that even in the research labs we have to be ‘cold blooded’ businessmen…The first question is this: what is going to be the total world market for the technology?…The answer is, looking 15 years ahead, $1 trillion per year…we then have to ask what fraction of the market will belong to quantum computation…Now, how much could HP capture if it went after it very aggressively…[then] the question is if we could sell that 15 years from now that is the appropriate level of investment for that income stream?

Stan then incorporated development costs, risks and barriers and the time value of money to conclude:

…even when addressing a significant share of a $100 billion market that is 15 years in the future, the amount of money we should be spending now is about a million dollars per year.  In an industrial laboratory environment that’s about three researchers with their associated overhead costs.

Every engineering manager in the company knew how to play this calculation in reverse:  if we fund one full time engineer to pursue a new, untested idea, what is the possible income stream we would see from that research 3, 5, 8, or 15 years from now?  Many – maybe most – of the technical staff understood it, too. And yet, there were these L-C ideas that just never seemed to go away. A generation earlier Dick Hackborn had been a management champion for inkjet printing, a crazy, complicated way of spraying colored water on paper, that even today accounts for most of HP’s financial success. As far as I know Dick was not in the decision chain for printing solutions, but he was a very influential guy and his sponsorship swayed many opinions at the topmost levels of management.

So what was the Technology Council’s role in all of this?  The company was much bigger, and a consequence of size is a decreased reliance on individual opinion and an increased reliance on quantitative processes.  As a result new ideas needed to be accompanied by a business case analysis that supplied both the decision model and the critical financial and market parameters. The difficulty was that business managers were making decisions mainly about their markets and their risks which affects the starting point for Stan’s calculation and may dramatically underestimate the role that organizational barriers play in estimating the total risk.  The Technology Council was in a position to combine information from a number of business units and recalculate the business case.

Here’s one example. HP was at that time organized into four large business units:  one for personal computers, one for services, one for large servers, and another for printing.  The software in HP’s most expensive servers was a version of the original Unix developed at Bell Labs in the 1970’s called HP-UX.  It was one of the most important profit drivers for HP’s high performance business systems but it was under pressure from the high volume Microsoft-based market on one side and other Unix variants such as Linux, Solaris, and AIX on the so-called “value” side of the server market. The Printing Group also was in the software business, designing drivers and user interfaces for printers and scanners that were attached to personal computers and workgroup servers.  The focus of printing software was on the large and very profitable market for Microsoft-based PC’s, workstations, and servers.  By comparison, relatively few of the much more expensive HP-UX systems were sold.  The Printing Group did the Williams calculation and concluded that investing in software for HP-UX was not warranted.  The Server Group meanwhile was being starved for printing solutions.  Customers were asking for it.  Lack of HP-UX printing support meant lost sales, but HP-UX software developers would have needed engineering support from their colleagues in the Printing Group in order to make any headway.  Printing did not see enough downstream revenue to justify such an investment.

A Loose Cannon proposed that my office should fund a cross-business initiative in HP-UX printing solutions.  When the Technology Council looked at the opportunities that were being lost, it was clear that even a modest investment would pay off in the very near term.  Although we didn’t realize it at the time, it turned out that HP’s investment in Linux would quickly  take hold in the marketplace, so the investment in HP-UX printing had a big impact on that market as well.

There were worlds smashing into each other all over the place in those days, and there were two organizational decisions that made a difference.  The first was Carly Fiorina’s decision to make the CTO a member of  the company’s Executive Council – the half-dozen executives who ran the company.  This added a technology voice to the most significant decisions made at HP. Having a seat at the table is important when worlds collide, and I will give many examples of this in later posts. The second was the decision to charter the senior technologists in the company to spend an entire day every quarter looking beyond their own business plans for new technologies and products that would have been dropped or gone unnoticed because they had not survived Stan Williams’ cold blooded calculation within a business silo.

Many other developments grew out of these Loose Cannon discussions including HP’s aggressive entry into open source software, supercomputing, and commercial printing.  Successfully bringing Loose Cannons into the fold really requires you to squarely face  two important issues.  The first concerns the role that organizational barriers play in affecting overall technology strategies, The second is why technologists don’t more often have a meaningful seat at the table in executive suites and boardrooms. More on how to deal with these issues later, but I will give you a hint right now: there are no clean solutions because worlds are in collision.

I arrived at HP long after Steve Wozniak sent his letter asking for permission to commercialize “hobbyist” computers (see my last post Proposition 13 and Innovation).  If  he and I had overlapped I wonder if he would have been one of my Loose Cannons and whether his letter would have been needed.

---

[1] “Nanocircuitry, Defect Tolerance and Quantum Computing: Architectural and Manufacturing Considerations” by R. Stanley Williams in Quantum Computing and Communications edited by Michael Brooks, Springer 1999.