IP: Stewart Baker's Summary of the Paris Encryption Summit

[Dave Farber <farber () central cis upenn edu>]

Date: Sat, 03 Feb 96 21:32:29 EST
From: "Stewart Baker" <sbaker () mail steptoe com>
To: farber () central cis upenn edu


    SUMMARY REPORT ON THE
     OECD AD HOC MEETING OF EXPERTS ON CRYPTOGRAPHY
     
     by
     
     Stewart A. Baker
     Steptoe & Johnson
     Washington, DC
     sbaker () steptoe com
     
     The OECD's ad hoc meeting of experts on cryptography was the 
     brainchild of U.S.  policymakers.  Export controls on 
     encryption have increasingly been attacked as unworkable by 
     U.S. software and hardware producers, who see a major market 
     for security on the global information infrastructure.  This 
     need, they argue, will be met by foreign producers if U.S. 
     export controls are kept in place.  Many companies in the 
     software business have also attacked the latest 
     Administration proposal allowing the export of strong 
     encryption only if it incorporates some form of key escrow.  
     These companies question the international demand for key 
     escrow.
     
     The likely U.S. purpose in calling for the OECD meeting was 
     to show that other nations are or soon will be inclined to 
     favor key escrow in order to avoid the problems that 
     criminal use of encryption will pose.  From the U.S. point 
     of view, the meeting was an opportunity to raise the 
     consciousness of other governments about the problem of 
     uncontrolled encryption while at the same time demonstrating 
     to U.S. industry that defeating U.S. export controls would 
     not open the door to a vast market for unescrowed encryption 
     but would instead spark new and perhaps inconsistent local 
     regulation of encryption.
     
     If that was the purpose of the meeting, it was a qualified 
     success.  It was not a complete success, because several 
     governments expressed grave doubts about the U.S. effort to 
     control encryption technology.  Most prominent among the 
     doubters were the Scandinavian countries.  Japan also showed 
     little interest in controlling encryption;  it seemed more 
     concerned about catching up in this former defense 
     technology now that its commercial possibilities were 
     growing.  
     
     Other governments, in contrast, were supportive of some kind 
     of escrow, though they disliked that term and preferred to 
     speak of "trusted third party" approaches to key storage and 
     recovery.  The European Union, the United Kingdom, and 
     France clearly favor the development of trusted third party 
     encryption systems.  Other countries also said favorable 
     things about trusted third parties.  But that term is 
     deliberately ambiguous.  It mixes together a wide variety of 
     "trust" services for users of computer networks.  Some 
     services, such as maintaining a register of digital 
     signatures or providing digital timestamps, do indeed 
     require trust but are quite uncontroversial.  At its most 
     minimalist, support for "trusted third party" encryption 
     might simply mean that governments will set standards that 
     allow companies performing uncontroversial  "trust" services 
     to also perform private key escrow when asked to do so by 
     users of encryption systems.  Such an approach is unlikely 
     to make escrowed encryption the dominant form of secure 
     computer network communication.
     
     But at least some European governments plainly mean to do 
     more than that under the heading of trusted third party 
     encryption.  Both Italy and the Netherlands have recently 
     considered legislation to regulate encryption directly in 
     the fashion of the French.  The UK is also disinclined to 
     see the spread of uncontrolled encryption within its 
     borders.  While these governments now seem unlikely to adopt 
     French-style encryption regimes, they are clearly intrigued 
     at the thought that, with government's thumb on the scale, 
     European telecom and computer companies might be willing to 
     adopt trusted third party encryption even without a direct 
     government mandate.
     
     The good news for U.S. policymakers is that many European 
     governments are clearly interested in doing something to 
     encourage key escrow encryption, and Australia and Canada 
     are likely to follow if a consensus in favor of key escrow 
     emerges.  This proposal for international concensus is bound 
     to cause some of the most vocally anti-escrow U.S. companies 
     at least a moment of self-doubt.  
     
     The bad news for U.S. policymakers is that there is little 
     appetite in Europe (let alone Japan) for direct regulation 
     of the encryption market (even the French are showing more 
     flexibility in enforcing their law).  And some European 
     governments' commitment to trusted third party encryption 
     may not go beyond saying nice things about it while waiting 
     to see what the market does.
     
     For a variety of reasons, the OECD is likely to be drawn 
     into the process of making international encryption policy.  
     The U.S. was generally pleased with the warmth -- if not the 
     ambiguity -- of the international praise for trusted third 
     party encryption, and it hopes to build a stronger 
     international consensus for such encryption.  Industry, 
     particularly U.S. industry, would rather see policy made in 
     the OECD than in a (presumptively more protectionist) 
     European forum.  And the other OECD nations see that forum 
     as a good place to moderate unilateral U.S. policies, such 
     as the current requirement that keys be escrowed in the 
     United States.  Thus all of the participants have something 
     to gain from continuing the dialogue in the OECD.
     
     ----------------------
     
     A more detailed description of the conference will be posted 
     shortly to my law firm's web page.  To see if it's up, go to 
     "http://www.us.net/~steptoe/welcome.htm"; and look under "Law 
     and the Net"