John Edstrom edstrom at
Sun Apr 28 15:03:19 EST 1996

In article <4lrgsa$3krs at>,
	hrubin at (Herman Rubin) writes:
>In article <4lo808$821 at>,


>Do you not think that there will be monitoring of the use of procedures?

Of course, but by whom and where will the money come from to do it?  And not
just procedures but attention needs to be paid to the results and their
interpretation.  I can cook up a phoney research paper that claims plausible
results were obtained from impeccable procedures.  It could fool experts but
it would still be a lie.  Such a fraud could go undetected without others
actively involved in research in the field who could test those results and
their consequences.  Just monitoring reports of procedures won't do the job.
And that is a deliberate fraud.  Well-intentioned people will often honestly
misslead themselves and others by simple error or loss objectivity.

>I doubt that there will be little of the current wrangling on tobacco if
>both sides, the government and the companies, had to be honest.  All it
>would have taken to prove in a civil case that tobacco was addictive would
>be to have the people hooked testify.  This would not prevent them from

Oh come on, if people thought they could get money by claiming damages from
drinking milk they would.  There'd be people lined up around the block ready
to swear on a Bible that they were addicted to milk and suffer its balefull
influences because of the cynical machinations of a rapacious dairy industry.
The self-interest, greed and dishonesty of the individual is neither greater
nor less than that of corporations.  What is needed is objective and rational
analysis of empirical data.  Who's to pay?


>>>The major funding of basic research before WWII was from university
>>>endowments.  Consider that a many major private  universities would have
>>>to add several BILLION to their endowments to fund what the federal
>>>government provides, although the actual amount could be reduced by
>>>about a factor of three to remove the administrative costs.
>>How do you figure this?  Much of the overhead costs come from the schools now
>>as it is.
>Huh?  Grant proposals in the mathematical sciences typically have an 
>overhead charge which is more than half of the salary charge.  

Sure, the same as in the life sciences and physics (and probably other
disciplines).  [For those not familiar with grant funding in the US a
university will deduct some amount from the grant of a researcher.  The
deduction is rationalized on the premis that conducting research on campus
entails costs to the university.  These costs include things from mundane
matters like water and power, building construction and mainenance, bookeeping
and purchasing, to more research-related costs like library books and journal
subscriptions  and other things.  The amount deducted varies from university to
university.  Typical values are from 60-110% of the grant.  This means that if
you want to do $100,000 worth of research you need a $200,000 grant, about
half of which goes to the university.]  These overhead costs are typicly
charged to all grants, whether they come from NIH or the March of Dimes so I
don't see how changing the source of the funds will make a fundamental change
in the amount that goes to reasearch and the amount that goes to overhead.

> If the schools want more research per dollar why don't they reduce
>>their admistration costs now?  These costs are common to funds from government
>>agencies and non-government sources so the same "inefficiencies" will afflict
>>research no matter who funds it.
>They cannot reduce their administrative costs.  Much of this is mandated
>by the government.  There are endless reports, proposals, audits, etc.
>There has to be so many accountant types for so much research activity.
>There was essentially none of this before WWII.  There was no "time
>accounting" for 30 years afterward.  If there was a chemist on the faculty,
>it was assumed that he or she would be using equipment, needing graduate
>assistants, chemicals, etc.; there was no grant process.  Now, everything
>has to come from a rigid account.  There is a separate account for every
>research grant, and an investigator is violating the law if a piece of
>equipment on one project is used for another without approval.  Switching
>of funds from computing to travel is easy, but between pay and equipment
>is a major hassle.
>>And if that wasn't the problem, just what massive administrative
>>inefficiencies are you complaining about?  Where exactly do you think the
>>money is being wasted?
>There would not be a "research office", with its staff.  There would not
>be all the accounting we now have.  Every grant proposal has to be approved
>by several administrators, and its budget has to satisfy several legal
>requirements, which keep changing.  There was a recent situation in which
>NSF changed their page limitations after proposals went in; there was a
>fight to allow the universities to resubmit.

I agree with part of this.  The bureaucratic aspects are often counter
productive and maladaptive and need reform.  But they are not there without
reason.  Accountability is necessary.  I think you agree.  The problem is to
get a system that gives reasonable assurances that money will be spent
responsibly while still leaving the researcher free to allocate resources as
he or she feels are most appropriate for prevailing circumstances at a given
time.  But I think there is a big difference between saying something should
be changed and saying it should be abolished.

It should also be kept in mind that your proposed idea, funding research at
univerities from their own research budgets, has its own problems that need to
be considered.  Which is to say that while national level funding has its
problems, locally funded research won't be all beer and skittles.  For
example, every department I've ever been in has had its parochial politics of
one flavor or another.  The influences of people within institutions is
typically stronger than that person's influence outside the institution.
Independent, portable funds, such as come from federal grants, give a degree
of independence that isn't found with funds that are tied to a particular
institution and are dispensed by some malevolent old fossil with a Napoleon

I agree with you that there are real and serious problems with the way
scientific funding is set up today, but I don't believe that your proposed
alternative is superior to what we have now.  It is not less vulnerable to
abuse, it is not immune from creeping inefficiencies, and it is almost
certainly liable to have problems no one has anticipated yet.

>Herman Rubin, Dept. of Statistics, Purdue Univ., West Lafayette IN47907-1399
>hrubin at	 Phone: (317)494-6054	FAX: (317)494-0558

 John Edstrom | edstrom @
 "Lurker" at BioMOO (

 Hatfield Marine Science Center
 2030 S. Marine Science Drive
 Newport, Oregon     97365-5296
 wk: (541) 867 0197
 fx: (541) 867 0138

More information about the Bioforum mailing list