4 ms·

The notion ghat “the government” is ‘bad’ at spurring innovation in high-risk scenarios is both common and flawed. It was NASA that sent men to the moon and, mo

by qubex 8y ago

The notion ghat “the government” is ‘bad’ at spurring innovation in high-risk scenarios is both common and flawed. It was NASA that sent men to the moon and, more pertinently, it was ARPA that shouldered the risks upon which the predecessor of the Internet (and packet-routing networks generally) was built, as well as (to a lesser degree) the microprocessor. I’m not quite sure why this is such a persistent impression, parroted by folks who, if they thought it through properly themselves, would reach the opposite (correct) conclusion.

aaavl2821 8y ago
I think the concern is not that the government can't spur r&d, but publicly funded r&d would be less efficient than private capital Back in the 1960s no other organization but the government had enough money to fund something as ambitious as the space program. And the government spends a ton of defense research and always has. Sheer volume of capital can get things done, but that doesn't mean it is efficient The belief is that private markets incentives more efficient research than public funding. So if there is enough private capital to support some r&d goal, it is preferred to fund it privately vs publicly I'm sure there's research on this but as someone who has worked on r&d at small companies, large companies and with government funded grants, there's little doubt in my mind what groups are more efficient. If you look at r&d productivity at big pharma companies vs small, small companies are almost an order of magnitude better. And out of 47 new drugs approved last year, 40% were developed by small companies, 30-35% by big companies, and 1 from a public private partnership (the rest were mid sized companies). That's despite small companies getting an order of magnitude less funding than big ones, and half as much funding as NIH
- philipkglass 8y ago
  Does that 40% figure refer to taking drugs all the way through clinical trials and FDA approval, or does it also include e.g. drugs that small companies brought through earlier stages of the pipeline before well-capitalized BigCorp acquired/partnered to finish approval? My impression is that drug development costs are heavily weighted toward the later phase clinical trials, and that it’s rare for a startup to complete them without help/acquisition from one of the big incumbents. I could be wrong about this. My lay perspective is that of someone who used to be a (non-pharma) chemist and who enjoys reading Derek Lowe. But if I’m right, and small companies usually can’t finish fully independently, that suggests a role for government drug development: funding the clinical trials and completing approval. Whether a government agency or GSK provides that funding doesn’t seem to much affect innovation of small companies in earlier stages.
  - aaavl2821 8y ago
    I'll have to check my data. You are right re late stage studies being most expensive component of drug dev cost in terms of direct costs, but when you account for cost of failure, early stage dev is actually more expensive. Phase 2 failure (drugs that work in the lab and animals not working in people) is the single biggest cost driver. That cost is linked highly to the difficulty of biology and biochemistry (the stuff Derek Lowe often talks about :) and it isn't clear to me how more public funding would address this Your proposal of having govt fund later stage r&d makes more sense to me that replacing IP altogether but i still think the current situation would be better at getting new drugs approved more cheaply Reason being is that early stage innovation is tightly linked to what assets pharma companies want. Allocation of venture capital by therapeutic area is very close to allocation of approved drugs by therapeutic area and to big pharma acquisitions of small biotechs by therapeutic area If government chose which drugs to fund at later stages, that would probably resemble the current NIH grant approval procedures which are fraught with politics and bureaucracy. There are some NIH / govt initiatives now to fund clinical studies (companies I've worked with have applied to two, CIRM and an NIH funded diabetes research group) and the application processes were brutal and took years. And I don't think those initiatives have yielded much in the way of approved drugs
    - philipkglass 8y ago
      I don’t think that government taking drugs through clinical trials would significantly reduce development costs. I think it could reduce all-in costs for drug buyers if some new drugs were “born generic” in terms of IP restrictions. Everybody’s revenue is somebody else’s cost. The idea is to shrink the cost/revenue for new drugs that government takes over the finish line, whereas private investment toward the same goal needs to recover R&D costs plus make a profit. I don’t think that lower drug prices is a panacea for health care costs in the US! Drug spending is a small slice of an oversized pie. But that pie is so big that even small improvements at the margin could save patients billions - mostly unnoticeable at the individual level, but worthwhile considered at the macro scale.
      - aaavl2821 8y ago
- refurb 8y ago
  Agreed. Don’t get me wrong, I think the gov’t should still play a role in funding “big ideas” that the market won’t. Antibiotic development is a good one. The other thing to keep in mind are all the failures. By pushing R&D to private companies, the tens of billions spent on Alzheimer’s drugs, of which every single one has failed, didn’t come from the pockets of tax payers, rather they came from private investors.

by qubex 8y ago

aaavl2821 8y ago

I think the concern is not that the government can't spur r&d, but publicly funded r&d would be less efficient than private capital Back in the 1960s no other organization but the government had enough money to fund something as ambitious as the space program. And the government spends a ton of defense research and always has. Sheer volume of capital can get things done, but that doesn't mean it is efficient The belief is that private markets incentives more efficient research than public funding. So if there is enough private capital to support some r&d goal, it is preferred to fund it privately vs publicly I'm sure there's research on this but as someone who has worked on r&d at small companies, large companies and with government funded grants, there's little doubt in my mind what groups are more efficient. If you look at r&d productivity at big pharma companies vs small, small companies are almost an order of magnitude better. And out of 47 new drugs approved last year, 40% were developed by small companies, 30-35% by big companies, and 1 from a public private partnership (the rest were mid sized companies). That's despite small companies getting an order of magnitude less funding than big ones, and half as much funding as NIH

philipkglass 8y ago
Does that 40% figure refer to taking drugs all the way through clinical trials and FDA approval, or does it also include e.g. drugs that small companies brought through earlier stages of the pipeline before well-capitalized BigCorp acquired/partnered to finish approval? My impression is that drug development costs are heavily weighted toward the later phase clinical trials, and that it’s rare for a startup to complete them without help/acquisition from one of the big incumbents. I could be wrong about this. My lay perspective is that of someone who used to be a (non-pharma) chemist and who enjoys reading Derek Lowe. But if I’m right, and small companies usually can’t finish fully independently, that suggests a role for government drug development: funding the clinical trials and completing approval. Whether a government agency or GSK provides that funding doesn’t seem to much affect innovation of small companies in earlier stages.
- aaavl2821 8y ago
  I'll have to check my data. You are right re late stage studies being most expensive component of drug dev cost in terms of direct costs, but when you account for cost of failure, early stage dev is actually more expensive. Phase 2 failure (drugs that work in the lab and animals not working in people) is the single biggest cost driver. That cost is linked highly to the difficulty of biology and biochemistry (the stuff Derek Lowe often talks about :) and it isn't clear to me how more public funding would address this Your proposal of having govt fund later stage r&d makes more sense to me that replacing IP altogether but i still think the current situation would be better at getting new drugs approved more cheaply Reason being is that early stage innovation is tightly linked to what assets pharma companies want. Allocation of venture capital by therapeutic area is very close to allocation of approved drugs by therapeutic area and to big pharma acquisitions of small biotechs by therapeutic area If government chose which drugs to fund at later stages, that would probably resemble the current NIH grant approval procedures which are fraught with politics and bureaucracy. There are some NIH / govt initiatives now to fund clinical studies (companies I've worked with have applied to two, CIRM and an NIH funded diabetes research group) and the application processes were brutal and took years. And I don't think those initiatives have yielded much in the way of approved drugs
  - philipkglass 8y ago
    I don’t think that government taking drugs through clinical trials would significantly reduce development costs. I think it could reduce all-in costs for drug buyers if some new drugs were “born generic” in terms of IP restrictions. Everybody’s revenue is somebody else’s cost. The idea is to shrink the cost/revenue for new drugs that government takes over the finish line, whereas private investment toward the same goal needs to recover R&D costs plus make a profit. I don’t think that lower drug prices is a panacea for health care costs in the US! Drug spending is a small slice of an oversized pie. But that pie is so big that even small improvements at the margin could save patients billions - mostly unnoticeable at the individual level, but worthwhile considered at the macro scale.
    - aaavl2821 8y ago
refurb 8y ago
Agreed. Don’t get me wrong, I think the gov’t should still play a role in funding “big ideas” that the market won’t. Antibiotic development is a good one. The other thing to keep in mind are all the failures. By pushing R&D to private companies, the tens of billions spent on Alzheimer’s drugs, of which every single one has failed, didn’t come from the pockets of tax payers, rather they came from private investors.