« "No Contracts" | Main | Hiring Committees 2015-2016 »

Monday, July 13, 2015

Line Drawing and Rulemaking

        As I suggested last post, I think that, as much as scholarship regarding tax law administration will benefit from relying on the (many) decades of administrative law scholarship, having fresh (tax law) eyes interacting with administrative law scholarship might also benefit the administrative law field.  To take one example, I have become fascinated by reading the administrative law scholarship regarding characterizing legislative rules, interpretive rules, and policy statements.  To put the matter simply, legislative rules must comply with notice and comment requirements.  Interpretive rules and policy statements need not.  Notice and comment is thought to integrate important values into the rulemaking process.  However, notice and comment is costly for agencies and, as a result, if it is required, agencies may avoid making rules.  Since it is very hard to distinguish between the three types of rules, many administrative law scholars worry that if too many things are characterized as legislative rules (thereby requiring notice and comment), agencies may reduce the amount of guidance that they issue. 

                One (among many) things that fascinates me about this issue is that, to the tax scholar (very much influenced by the decades of tax law and economics scholarship), this is a classic problem of linedrawing, and how to draw a line as efficiently as possible.  Essentially, we can reimagine the notice and comment requirements as a tax.  Rather than raising tax revenue (as an actual tax would), the notice and comment requirements serve a valuable goal.  However, if the tax (the application of the notice and comment requirements) is too high, people (in this case, agencies) will change their behavior to avoid it (in this case by not issuing guidance at all).  Changing behavior to avoid a tax is inefficient because (1) the tax isn’t raised, and (2) parties have changed their behavior from the optimal behavior they would have preferred in the non-tax world.  In the agency rulemaking context, the imposition of notice and comment requirements is inefficient when agencies simply avoid such requirements by not making rules because (1) the benefits of notice and comment procedures won’t be realized, and (2) the agency will not have issued the guidance it would have liked to issue absent the requirements.

                While it is fascinating to me just to see this problem through the linedrawing / efficiency lens, I think doing so may also yield helpful insights into the administrative law dilemma.  Inefficiencies are just a fact of life with taxes – when taxes are imposed, parties will shift their behavior to avoid them.  Similarly, the existence of notice and comment procedures will necessarily cause agencies to issue less guidance to avoid such procedures.  As a result, the very reduction in guidance shouldn’t cause hand-wringing.  On the other hand, the fact that agencies will inevitably change their behavior also shouldn’t be the end of the conversation.  Rather, the linedrawing scholarship in tax teaches that, while efficiency is not the only relevant criterion, all else equal, taxes should be imposed where behavior is least elastic.  Imposing tax where behavior is least elastic raises the most tax possible while engendering the least behavioral distortion.  Applying this principle to the context of agency rulemaking, then, notice and comment procedures should be imposed when the agency is least likely to change its behavior to avoid such requirements.  The question, then, is when is this likely to be the case?  I wonder whether empirical studies could help determine the likely elasticity of agencies’ responses to the imposition of notice and comment procedures in various situations.  While this may seem like quite a lot to ask as an empirical matter, many years of work in the tax context have revealed quite a lot regarding tax elasticities.  Perhaps merely posing the question at this point might yield some new opportunities for empirical study in the agency context.  At the very least, perhaps the linedrawing lens might be a new, helpful way to conceptualize a seemingly intractable problem in administrative law scholarship. 

Posted by Leigh Osofsky on July 13, 2015 at 11:03 AM in Tax | Permalink


Thanks for the thoughts, Susie. Great point about situations in which agencies following notice / comment may not get additional deference anyway (ie: with decisions not to enforce). In such situations, agencies have less incentive to bear the "tax." Perhaps this could help explain / justify courts' decisions to allow many nonenforcement policies to be made outside notice / comment.

Posted by: LZO | Jul 21, 2015 9:40:52 AM

Agreed that questions of whether agencies get more deference for notice/comment rules is tied up in this. If they clearly did perhaps more like a fee than a tax, or at least a fee to the extent that the value of extra deference matched up with the trouble of notice and comment. If additional deference not practically available for notice/comment rules (as perhaps it is not for decisions not to enforce) then I think the idea that agencies would avoid notice/comment rules is particularly strong.

Posted by: Susie Morse | Jul 20, 2015 6:09:25 PM

The analogy could run the other way to. The way it is supposed to work is that there should be less deference to informal rules, but a great many people don't believe that courts actually do. The fact that agencies don't use notice-and-comment rulemaking may suggest that the tax that is supposed to exist (less deference) doesn't actually exist in practice. Therefore, the courts' treatment of rules after notice and comment needs to change such that there is actually an incentive to use it.

Posted by: James | Jul 14, 2015 8:27:47 PM

Thanks for the comment, James. My thinking at this point was quite preliminary. But I think that the analogy would do more than pose a tax / don't tax dilemma. Basically, the elasticity research in the tax context shows that people do not always change their behavior to avoid taxes to the same degree. In different contexts, behavioral changes are greater / less. For instance, labor is relatively inelastic, meaning that people are less likely to change their labor behavior (working decisions / hours) to avoid a tax. In contrast, people are more likely to change their saving behavior (in terms of what and how they save, and the tax planning they will employ). Secondary wage earners are more elastic than primary wage earners. As a result, heavily taxing secondary earners (for instance, through the current marriage tax structure for some couples) can heavily discourage secondary earner employment. These are only some of the insights that have been revealed through extensive empirical work. The question would be what empirical insights we might gain if we started to view notice and comment (as a tax) through this lens. I could imagine, for instance, that an agency would be less likely to just avoid issuing guidance altogether when the rule was part of the administration's central platform. As a result, perhaps there might be ways to structure the notice and comment requirements so as to require notice and comment at least in situations in which there is likely to be a less elastic response. Administrative law scholars would understand much better than I would what the various possibilities / intuitions would be that would merit testing. But perhaps the tax analogy would motivate thinking about these possibilities and testing them. As you mentioned, there might also be substitutes for the notice and comment tax that would encourage agencies to use notice and comment -- for instance, giving less deference to interpretive rules and policy statements. There is a tax analogy for this insight as well. In tax, it makes sense to tax close substitutes for behavior when it is difficult to tax the behavior itself. For instance, imagine that we tax labor (we do). This encourages people to shift into leisure (undesirable behavioral shift - no tax raised, behavior has changed to less desirable behavior). We would like to tax leisure to prevent this shift, but cannot. So, we should potentially tax close substitutes to leisure (ie: going to the movies, etc.). Your comment of giving less deference to interpretive rules and policy statements similarly suggests a different kind of tax (less deference) that can help reinforce the tax we would like to impose on the primary behavior (issuing rules). I think there is a lot more here to think about -- and would be very excited to see how this could play out.

Posted by: LZO | Jul 14, 2015 9:15:26 AM

Interesting. It seems like an analogy worth pursing. There is no question in my mind that notice-and-comment rules act like a tax in a very literal sense. Complying with them costs money --- lawyers tend to be relatively expensive FTE --- and it seems that most agencies would prefer to spend their limited budgets doing "real work" rather than spending it on "legal hoops." The burden of rulemaking is an obstacle to completing the mission, so agencies look for ways around it.

So, does the analogy suggest an approach more sophisticated than tax/don't tax? Are there ways around the problem of taxing elastic behavior? Does it mean that we should be looking for ways to tax the inelastic behavior when there isn't notice and comment? It seems that providing less deference to informal rules is one such tax, but often not sufficient to result in notice-and-comment rulemaking.

Posted by: James | Jul 13, 2015 9:50:08 PM

Thanks for your thoughts, BDG. I am definitely operating from the assumption that the absence of new guidance is a cost. In so doing, I am picking up on the terms of the debate I have seen put forth by scholars wary of applying notice and comment rulemaking broadly (including Peter Strauss, among others). However, I think your point may underscore the utility of thinking of this through a tax / linedrawing / efficiency framework. For instance, if reducing guidance is a benefit, then more notice and comment could be seen as an optimal sort of tax, much like a tax on an externality like smoking. If the agency doesn't change its behavior, then we would get the benefit of notice and comment rulemaking. If it does, all the better because the reduction in behavior (guidance) would be a good thing. Again, I was firmly operating from the assumption that reducing guidance would, in fact, be bad. But, I think the framework may be helpful in forcing us to think through what really is a cost and what really is a benefit. In any event, I haven't seen any empirical work along the lines I suggest and would be interested to know if there is any I am missing, and whether this framework might be helpful to nudge it forward.

Posted by: LZO | Jul 13, 2015 4:15:47 PM

Interesting! But is there a neutral way to decide whether the absence of new guidance is a cost (as in your framework) or actually part of the point? A cynic might say that the D.C.'s circuit's approach has generally reflected the view that all new guidance is bad guidance. This would be akin to the libertarian-textualist position that increasing the costs of legislation is always good (notwithstanding the critique, never convincingly rebutted in my view, that most legislation actually amends and may even repeal old law).

Posted by: BDG | Jul 13, 2015 3:58:55 PM

The comments to this entry are closed.