The major differences between academic and industrial research

While there are real differences between academic and industrial research, there are a lot of misconceptions. The three most common misconceptions I've heard are:

1. "They [industrial researchers] have tons of money, so they can get whatever equipment they want."
2. "They [industrial researchers] lack curiosity and only care about money."
3. "We have to spend more on education to improve our economy."

The role of money

The perception by academics that industrial researchers have a lot of money appears to come from the facts that industrial researchers have higher salaries then their academic counterparts with comparable experience, and industrial labs often have much more modern state-of-the-art instruments. This difference is really due to how money is used, not how much is available, as determined by:

1. salaries
2. who pays those salaries
3. where the money comes from
4. priorities

Salaries

Science graduate students typically earn stipends between $20,000 to $30,000 plus some tuition reimbursement, and until recently typically very minimal healthcare. If they get vacation, it is usually by arrangement only, with Christmas and federal holidays being the exceptions. Over their tenure as graduate students, they may get 1 or 2 milestone pay raises of $1,000 to $3,000 per year. Currently the average science graduate student is taking 6.5 years to finish their PhDs.

The stipends paid to academic postdocs changed this in 2014, and are usually set by NIH Kirschstein-NRSA awards (Postdoc pay). While an institution may pay an additional amount on top of the NIH Kirschstein-NRSA awards, they are generally very close. Based on the awards, a postdoc who just earned their PhD now make $42,000/ years, and after 7 years they reach a maximum of $55,272/year. Until the Affordable Care Act went into effect, the institution they worked at controlled how much, if any, medical insurance the postdoc received.

Now consider industrial researchers who, regardless of their level, have vacation/sick leave, full medical (with co-pay), and some type of retirement plan. A recent BA level graduate can expect to earn between $40,000-$50,000/year, and after 5-7 years should be making between $55,000 - $65,000/year. A new PhD typically earns between $75,000 and $85,000/year, though more is certainly possible. Please note, if an individual is working in an analytical lab or manufacturing setting their pay will be a bit less.

Who pays those salaries, and where does the money come from?

Within at least the last 30 years, the majority of science graduate students were paid from their department's budget in exchange for working as teaching assistants (TA). Some did receive their stipends from their adviser's research funds (RA) and very few won fellowships. This situation is changing as universities face tighter budgets and begin to demand that professors use more of their research grants to pay some of their graduate student's stipends.

Academic postdoc stipends have always come from either their professor's research grant or from a fellowship they won.

Industrial researchers are ultimately paid by their employer's shareholders from the employer's profits, or from the employer's investors in the case of start-ups.

Priorities

Again this is changing as universities face tighter budgets, but historically universities sought to increase the size of their programs. Larger graduate programs has meant more revenue (at public schools, the states pay some of the stipends), and more prestige (which attracts more undergraduates, who directly bring more revenue).

For business, the first priority is to earn more money than is spent. Once break-even is achieved other values may temper a business' priorities, but profit is always a top priority.

How do these factors lead to the equipment differences seen between academia and industry?

Since labor is typically a businesses largest expense, so minimizing its cost has the largest effect on a business' profitability. There are two basic ways of doing it: have a large staff using low tech equipment doing labor intensive work while paying low salaries and demanding long work hours; or have a small but well paid staff using highly automated equipment to do most of the work that would otherwise be labor intensive. There are pro's and cons to both approaches. The low tech/low salary/large staff approach requires a small initial investment but growth becomes difficult due to scaling issues and small profit margins. The high wage/high tech approach requires a large initial investment but considering modern automated scientific instrumentation often affords a 5 fold or greater force multiplier, this approach has a lower net overhead and often is easy to scale.

I've worked at companies that use both approaches and experience tells me that the large staff/low tech/low wage approach is the wrong one within the US - somewhere in the world, there will always be a place where the wages are lower. While the high wage/high tech approach does not guarantee success it helps - with a force multiplier of 5, one can pay an employee $80,000-$100,000/yr to compete against a low tech group of 5 employees earning $20,000/yr plus a supervisor (at $20,000+/yr). For that reason, the most successful science-based business in the US have chosen to go the high tech route.

Ironically, the low wage/low tech approach is largely the approach universities use with their graduate students, and with the large staffs they seldom have the money to get the latest technology... The other side of this is due to the historical fact that the graduate students were mostly paid from department funds and not their adviser's research funds, so their adviser's felt little economic pressure to adopt the latest technology. All of this is changing (though slowly) due to increasingly tight university budgets, and it is a very painful process.

Do industrial researchers really lack the curiosity of their academic counterparts?

The simple answer is no. It may appear that way because industrial researchers are focused on achieving particular goals at the lowest cost possible. To understand this better, let's examine the "research tree" below.

A research tree. One starts at the green dot, the arrows are experiments and the remaining dots as results.

While both academic and industrial research have objectives (the blue dot), academic researchers are encouraged to follow all interesting leads including the ones not relevant to their objective (all paths). Industrial researchers can't afford to do that because investors want income generating results and won't provide more money unless that is happening. In practice, the industrial researcher follows the most promising results only (the yellow arrows). Then if time, money or a need for a backup plan exist, good but less promising results will be followed. When an industrial researcher discovers something that is interesting but not relevant to their objective (yellow dots), they are suppose to report the result to management who then decides whether or not to do further follow-up.

As a side note, often there really are multiple ways to reach one research objective (as shown by the three paths to the blue dot). The prime example of this comes from the pharmaceutical industry where there are many drugs to treat the same medical condition.

Open-ended research vs. Profit Driven Research

Another reason why an academic might think industrial researchers lack curiosity has to do with the fact that academic research is open-ended whereas industrial research is not. Again the majority of academic research is basic research where the principle product is knowledge. As long as an academic researcher does what they said they were going to do in their grant proposals and write papers, they can apply for more grants and will never run out of questions to ask. This is true even if they don't reach their objective, as long as they produce new knowledge. This encourages them to forge ahead despite poor results, the "never give-up" attitude.

That is very different from industrial research which is profit driven. Well disciplined companies shutdown research projects when they fail to meet objective goals, all on the principle of "don't throw good money after bad." It is common for companies to talk about Go/No-Go project goals (objective standards of success, such as "produce a drug candidate with a 100 nanomolar IC50") which are evaluated once every three to six months. If a project fails to meet a Go/No-Go goal, the company kills the project. Sometimes if a project is close to its Go/No-Go goal, it might be put on probation for one or two cycles then if it still fails the company kills the project. Do industrial researchers get emotionally invested in their projects and push on well past the point it should be obvious to all that no real progress is being made? Yes, of course, researchers are only human but if a company allows that to happen too often and for too long, then the company will fail.

What is the role of education in stimulating economic growth?

Over the years I've heard a lot of versions of "the economy is bad, we need to invest in education," implying that education will make the economy grow. Unfortunately that line of reasoning is a non sequitur. I've also heard several versions of "we need to stimulate innovation, so let's invest in academic research," which is also a non sequitur. Hopefully by now, the second one is easy to understand: academic research primarily produces knowledge (that is why they write all those papers), not new products (innovation).

Besides the fact that education is expensive, the easiest way to explain why educational investments don't directly lead to economic growth is with a gardening analogy. In order for a farmer to grow tomatoes, he needs three things: fertilizer, tomato seeds, and water. While one might argue over what variety of tomato seeds to plant, one has to plant tomato seeds to get tomatoes. Nothing else will do.

To grow a business, one needs three things: fertilizer, business seeds, and water. Paying customers are the water. Ideas and infrastructure to produce products are the business seeds. Capital, a skilled workforce, and knowledge are the fertilizer. Education produces a skilled workforce and knowledge (from basic research). It is true that if one has a pile of fertilizer laying around, occasionally wind blown seeds will sprout in it but it is not an efficacious way of growing tomatoes. In much the same way, university research occasionally produces a spin-off business.

Oregon has good schools that train far more scientists then we have jobs for. Investing more in education will only make that problem worse, like drilling a hole in the bottom of a leaky boat while it is in the middle of a lake. There are a lot of good reasons to invest in education, but it is also time for Oregon to invest in infrastructure for science-based start-ups.  

Kirschstein-NRSA awards

Kirschstein-NRSA awards

Kirschstein-NRSA award