In Praise of Secrecy

Concluding that:

Approaching the problem under the premise that fuller transparency is always desirable may not be the right place to start.

an article On the Social Cost of Transparency in Monetary Economies from the St Louis Fed explains why secrecy and non-disclosure of data may be advantageous.

For an asset economy then, the prescription of “full transparency” is not generally warranted.

After many pages of mathematics, their argument is summarized:

In competitive economies, the disclosure of high-frequency information unrelated to an asset’s “long-run fundamentals” may be detrimental to economic welfare when claims to such assets serve as high-velocity payment instruments. The equilibrium price of a liquid asset is excessively volatile when asset prices capitalize all information.

If we translate this into a more familiar example, such as climate data and code, it would seem to argue that disclosure of all data and code may be detrimental to the value of the research to the author of an article, the university, the journal, coauthors etc.

Attempting to optimize the value of the research explains non-disclosure. This is because — in an econocentric view — researchers with an “equity share” in products of research may use these “tokens” as a medium of exchange.

For example, where co-authorship might be exchanged for access to private data or code, this enhances promotion or further grants. But the value of the researcher’s equity share becomes “excessively volatile” when all available information is factored into the asset, i.e. when it becomes known that the code is a mess, the data set is composed of only 7 trees, etc., its value diminishes. This information is some-what divorced from the “long-run fundamental value” of the research program.

Suppressing this high-frequency information flow stabilizes the short-run expected return of a monetary instrument (around its long-run fundamentals), so that consumers are never caught short of “cash” in bad news events.

Therefore, the excuse for non-responsiveness to FOI requests used in the Nature editorial Climatologists under pressure does not present the full picture.

If there are benefits to the e-mail theft, one is to highlight yet again the harassment that denialists inflict on some climate-change researchers, often in the form of endless, time-consuming demands for information under the US and UK Freedom of Information Acts. Governments and institutions need to provide tangible assistance for researchers facing such a burden.

Sad to see Nature perpetuating the ‘moral hazard’ of name-calling. But apart from that, they think the problem with disclosure is the uncompensated distribution cost, but not the uncompensated loss of asset value. Taking the analysis of David Andolfatto even further, one could argue that climatologists should be compensated for the loss of value of their equity share in the data and code, as a result of full disclosure.

Its not as crazy as it sounds. Fund the private research, then fund the disclosure of the products separately. If the research project fails, there is no cost to the public from the additional archival premium. Generally, financial incentives are needed and provide the additional push to complement the strategy of badgering people to do the right thing. The carrot and stick.

Advertisements

0 thoughts on “In Praise of Secrecy

  1. This is an interesting approach if you neglect that econometric models have a disaster failure rate and thus do not serve as good illustrations of transparency. In the present context, the problem with the analogy is that transparency of financial data can change the I.P. and wealth of others who sometimes show irrational behavior , but it cannot effectively change the climate.We analyse in the frames of our experience, so I ask if premature discolsure of our past mineral exploration results would have been good or bad. Then good or bad for whom, competitors or economist/investors or the national economy? (I say premature release because the law required the eventual release, so perhaps its is best to ask if release was of overall benefit or detriment).The overall conclusion to this thought exercise is that early release enhances the likelihood of error detection and correction, in both the data and the methodology. The disadvantage is a loss of incentive for the leaders to maintain front running, as free riders hop on the wagon.That front runing has a true value only if it works for the common good. If the common good can be identified (as in more original wealth creation from new mineral discoveries), then front runners have a place. If it cannot be identified (as in the difficulty of changing the climate as opposed to gossiping about it) then transparency is the logical preference. If it is a common bad with perverse intent, then transparency is essential. Discerning if it is good, unidentifiable or bad comes back to economics and prospering by reward. The punters eventually find out about the duds.A case can be made for the progressive and prompt enforcement of standards (such as for data transmission or methods of ore analysis or certification of laboratories). The failure shown by several cases in Climategate would not have happened if standards for transparency were policed. Often standards were in place, as journal or association or funding policy, but often they were swept beneath the tablecloth. That is where the real lesson sits.If you wish to publish without releasing your code and data, then don't publish. Heard that before?

  2. My understanding of it, that in the case of mineral exploration, the front runners might suffer a cash crisis, while the market is fluctuating around the long term expected value, and the secrecy reduces the volatility.Standards and enforcement are not working. But if there was a wall between those papers that have data and code available, and those that don't, and a higher value placed on the latter, maybe?

  3. This is an interesting approach if you neglect that econometric models have a disaster failure rate and thus do not serve as good illustrations of transparency. In the present context, the problem with the analogy is that transparency of financial data can change the I.P. and wealth of others who sometimes show irrational behavior , but it cannot effectively change the climate.

    We analyse in the frames of our experience, so I ask if premature discolsure of our past mineral exploration results would have been good or bad. Then good or bad for whom, competitors or economist/investors or the national economy? (I say premature release because the law required the eventual release, so perhaps its is best to ask if release was of overall benefit or detriment).

    The overall conclusion to this thought exercise is that early release enhances the likelihood of error detection and correction, in both the data and the methodology. The disadvantage is a loss of incentive for the leaders to maintain front running, as free riders hop on the wagon.

    That front runing has a true value only if it works for the common good. If the common good can be identified (as in more original wealth creation from new mineral discoveries), then front runners have a place. If it cannot be identified (as in the difficulty of changing the climate as opposed to gossiping about it) then transparency is the logical preference. If it is a common bad with perverse intent, then transparency is essential. Discerning if it is good, unidentifiable or bad comes back to economics and prospering by reward. The punters eventually find out about the duds.

    A case can be made for the progressive and prompt enforcement of standards (such as for data transmission or methods of ore analysis or certification of laboratories). The failure shown by several cases in Climategate would not have happened if standards for transparency were policed. Often standards were in place, as journal or association or funding policy, but often they were swept beneath the tablecloth. That is where the real lesson sits.

    If you wish to publish without releasing your code and data, then don’t publish. Heard that before?

    • My understanding of it, that in the case of mineral exploration, the front runners might suffer a cash crisis, while the market is fluctuating around the long term expected value, and the secrecy reduces the volatility.

      Standards and enforcement are not working. But if there was a wall between those papers that have data and code available, and those that don’t, and a higher value placed on the latter, maybe?

  4. This is an interesting approach if you neglect that econometric models have a disaster failure rate and thus do not serve as good illustrations of transparency. In the present context, the problem with the analogy is that transparency of financial data can change the I.P. and wealth of others who sometimes show irrational behavior , but it cannot effectively change the climate.We analyse in the frames of our experience, so I ask if premature discolsure of our past mineral exploration results would have been good or bad. Then good or bad for whom, competitors or economist/investors or the national economy? (I say premature release because the law required the eventual release, so perhaps its is best to ask if release was of overall benefit or detriment).The overall conclusion to this thought exercise is that early release enhances the likelihood of error detection and correction, in both the data and the methodology. The disadvantage is a loss of incentive for the leaders to maintain front running, as free riders hop on the wagon.That front runing has a true value only if it works for the common good. If the common good can be identified (as in more original wealth creation from new mineral discoveries), then front runners have a place. If it cannot be identified (as in the difficulty of changing the climate as opposed to gossiping about it) then transparency is the logical preference. If it is a common bad with perverse intent, then transparency is essential. Discerning if it is good, unidentifiable or bad comes back to economics and prospering by reward. The punters eventually find out about the duds.A case can be made for the progressive and prompt enforcement of standards (such as for data transmission or methods of ore analysis or certification of laboratories). The failure shown by several cases in Climategate would not have happened if standards for transparency were policed. Often standards were in place, as journal or association or funding policy, but often they were swept beneath the tablecloth. That is where the real lesson sits.If you wish to publish without releasing your code and data, then don't publish. Heard that before?

  5. My understanding of it, that in the case of mineral exploration, the front runners might suffer a cash crisis, while the market is fluctuating around the long term expected value, and the secrecy reduces the volatility.Standards and enforcement are not working. But if there was a wall between those papers that have data and code available, and those that don't, and a higher value placed on the latter, maybe?

  6. Pingback: zobacz tutaj

  7. Pingback: zobacz

  8. Pingback: rehabilitacja medyczna

  9. Pingback: darmowe anonse

  10. Pingback: wysokiezyski.blog4u.pl

  11. Pingback: polecam

  12. Pingback: polecam

  13. Pingback: montaz sufitów armstrong

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s