Category: Uncategorized

I have previously blogged (here and here) about the constitutional issues with the statute authorizing the governor to declare peacetime emergencies.  Minn. Stat. § 12.231. The best characterization of the legislature’s role under the statute is that it is constitutionally suspect. The legislative authority to terminate declarations longer than 30 days likely either gives the legislature executive powers or allows it to de facto enact legislation while circumventing the governor’s veto power. By compelling the governor to call special sessions, the statute also may unconstitutionally impinge on gubernatorial discretion to make those decisions.  This is all speculation on my part since I know of no Minnesota court decisions addressing these issues.

COVID-19 and Governor Walz’s declaration and executive orders have focused attention on the workings of this previously obscure statute. Naturally, the legislature considered changes to it based on that experience. The Senate passed a bill, S.F. No. 4519, that would have changed the legislature’s procedural role. The House did not take up the bill, so it died when the regular session ended.

So, would S.F. No. 4519’s changes have addressed the separation of powers issues with the current statute? On a strictly technical basis probably not, but as a practical matter it would. The bill prohibited the governor from extending a declaration beyond 30 days without each house of the legislature approving the extension by a majority vote. Technically, the legislature would be ratifying an executive action – not something exactly envisioned by the separation of powers. But the ratification process is almost the same as enacting the declaration as a law, since the governor’s declaration could be characterized as signing it into law or allowing it to become law without his signature. In this way it is similar to the procedure used for approving state employee contracts, also an odd hybrid arrangement of executive power shared with the legislature.

The current statutory procedure allows the legislature, if each house approves by majority vote, to terminate declarations of more than 30 days. That means deadlocks or ties allow the declaration to continue. By contrast, S.F. No. 4519, is about the equivalent of legislative enactment of declarations longer than 30 days.

It’s hard to know exactly what would have happened if S.F. No. 4519’s rules had applied during the current crisis. (As an aside, given the Senate vote it seems unlikely that this approach will find much favor in the DFL-controlled House in a special session, if it is revived. It looked to me like only one DFL senator voted for it.) It certainly seems possible that something more like the situation in Wisconsin (absent the need for a lawsuit) would have occurred after the end of the first 30-day period of Governor Walz’s declaration: that is, state-imposed restrictions on businesses and private behavior would have ended or been much more severely curtailed.

Requiring legislation to address an ongoing health crisis is likely to be more cumbersome and difficult, especially given a partisan split in control but even in its absence, compared to the governor acting alone. I would expect, based on my legislative experience, that the response would involve more horse-trading and give-and-take. The ability to make calibrated changes quickly in response to changing conditions would be hampered. Decisions would be more “political”- in particular, I would expect more exceptions and carve-outs needed to gain assent of a few or small groups of key legislators and/or to address local situations. That could be bad or good, depending upon how well you think the current arrangement is working.

As a final note, the bill includes a provision that addresses the possibility that a governor would evade the 30-day duration limit by allowing declarations to lapse and, then, after a short gap declaring a new emergency but to really address the old emergency.  This suggests to me a growing perception at least by some senators of the potential for gaming playing or bad faith by the administration. Not a good sign.

Governor Walz has extended his declaration of a peacetime (COVID-19) emergency by 30 days. Under the statute (Minn. Stat. § 12.31, subd. 2), the legislature can terminate the extension by majority vote of each house. As I noted previously, that authority likely is an unconstitutional usurpation of the governor’s executive power or unconstitutionally circumvents his authority to veto bills.

If Governor Walz extends the declaration when it expires on June 12^th, another constitutionally suspect provision of the emergency powers statute will be triggered. Because the legislature is unlikely to be in session on June 12^th (the regular session ends on May 18^th), a further extension would trigger the special session clause of the statute:

If the governor determines a need to extend the peacetime emergency declaration beyond 30 days and the legislature is not sitting in session, the governor must issue a call immediately convening both houses of the legislature. Minn. Stat. sec. 12.31, subd. 2.

Thus, the statute requires the governor to call a special session. A STRIB story uses a stronger formulation, implying the extension will “automatically call” the legislature into session.

The Constitution (article IV, section 12) requires the governor to call special sessions: “A special session of the legislature may be called by the governor on extraordinary occasions.” This language has been altered modestly by constitutional amendments but has always given the governor the discretion to decide both what an “extraordinary occasion” is that justifies calling a special session and whether to convene one. For example, the language of the original (1857) constitution provided: “He [the governor] may on extraordinary occasions convene both Houses of the Legislature.” The consistent use of “may” (albeit in the passive voice, which does create some ambiguity) strongly implies an intent to confer discretion to the governor to make the determination.

Reading the statute and constitution together raises the obvious question whether a statute can compel a governor to call a special session or if that contravenes the apparent gubernatorial discretion conferred by the constitution. As far as I am aware this issue has never been adjudicated in Minnesota or addressed by the Minnesota courts. (Caveat: I no longer have access to either a law library or legal research services like Westlaw, so I cannot research the question; the  following discussion is, thus, pretty much legal research-free and is little more than my speculation about the implications of the constitutional text. Normally one would look for other instances in which the court has opined about or ruled on the nature of the governor’s constitutionally granted powers to seek insight about the nature of the governor’s special session-convening power.)

The text of the Governor Walz’s order appears to concede that the statute can require him to call a special session, implying the requirement is constitutional, although I doubt much thought was put into the language:

Because the Legislature is currently sitting in regular session, there is no need to call a special session of the Legislature under Minnesota Statutes 2019, section 12.31, subdivision 2(b) Executive order (#20-53), p. 3.

The governor may simply decide for political or practical reasons to call a special session if he decides to extend the declaration. For example, doing so would avoid angering Republicans and could cast him as someone who seeks to work cooperatively with Republicans and the legislature generally. And it would avoid creating unnecessary legal doubts about his actions (another count or counts in a complaint challenging the declaration itself or more likely specific limits imposed under the declaration?). Since the DFL controls the House, it is highly unlikely his extension would be overturned by the legislature. However, I can easily imagine a desire to avoid having a potentially drawn out and open-ended special session occur just to satisfy the statute. And it would create a “custom and usage” style precedent regarding the ability of the legislature (via enactment of statute) to constrain or direct the governor’s authority/power to convene special sessions. (I have no idea whether Walz is the type of guy who is protective of executive prerogatives and powers. Most of his political experience was in the legislative branch, Congress.)

There are clear constitutional issues as to whether a statute can deprive the governor of discretion to determine when to convene special sessions of the legislature. For example, I doubt anyone would conclude that a statute requiring the governor to call a special session upon written request of the Speaker of the House and President of the Senate is constitutional. That would delegate a good portion of the power the constitution assigns to the governor. Section 12.31’s provision obviously presents a somewhat narrower and more difficult question.

I see three obvious general issues:

First, does the constitution confer exclusive or sole discretion on the governor to determine (1) whether an occasion is “extraordinary” and (2) then, to call a special session to address it? If either is the case, the statute is unconstitutional and cannot compel the governor to call a special session, authority that the constitution invests solely in the governor. The text provides little insight, as far as I can tell, into whether this was to be unfettered or exclusive power (i.e., not to be constrained or to be used as prescribed by law). The language simply authorizes the governor to convene sessions on “extraordinary occasions.”

There is no reason to assume the power is inherently executive. By its nature, the power has mixed elements of legislative and executive functions. It requires an executive decision (determining whether there is an “extraordinary occasion”) regarding the need for legislative consideration and/or action. In the constitutional debates, it was discussed whether to locate the power in the executive or legislative article.  The Debates and Proceedings of Minnesota Constitutional Convention, (Earle S. Goodrich, Territorial Printer, Pioneer and Democrat Office, 1857) (Democratic debate), p. 264 (later Governor Sibley observed the power to call special sessions belonged in both articles). It was placed in the executive branch article. Minn. Const. art. 5 § 4 (1857).  The 1974 amendment restructuring the constitution moved it to its current location in the legislative article. 1974 Minn. Laws ch. 409. As a result, little can probably be read into the text, its location in the constitution, or the original debates as to whether it is a uniquely executive or legislative power.

That likely will leave a court to its own devices, perceptions of the executive and legislative branches, and whatever predilections it has in that regard. My observation is that the courts have tended to look more favorably on the executive in recent interbranch disputes.

Second, if the constitution does not exclusively empower the governor to determine what is an “extraordinary occasion,” then does section 12.31, which statutorily deems any governor’s declaration of a peacetime emergency of more than 30 days to be such an occasion, meet that standard? An obvious argument is that the governor’s emergency declaration under the statute satisfies the constitution’s requirement that the governor determine whether there is an “extraordinary occasion” because the governor is the one who makes a (statutory) determination of a sufficient peacetime emergency. Thus, one can argue, a constitutional requirement that the governor make the determination is satisfied because under the statute he has made the determination. That argument’s potential flaw is that it is not obvious (at least to me) that all longer duration peacetime emergencies necessarily meet the criteria of “extraordinary” for purposes of convening of special sessions as envisioned by the constitution. One assumes that whatever those unstated constitutional criteria may be, they likely would focus on occasions that require changes in state law or appropriation of money (i.e., acts requiring passage of legislation). (Note: the debates in the constitutional convention make some passing, indirect references what would be justify calling a special session, in the context of debating whether they should be timed limited or not. See id. p. 265.) It is easy to imagine peacetime emergencies that can be addressed with exclusively executive action with no need to either change the law or to appropriate additional money. If Governor Walz were to decide it is unnecessary to convene a special session (i.e., that section 12.31’s requirement is not constitutional and there is no immediate need for any legislative action), that is precisely what he would be concluding.

It’s also possible that the supreme court would conclude that the constitution gives the governor sole discretion to make the determination and that reviewing that determination is non-justiciable, essentially a political question that the courts need to steer clear of or one that is not otherwise susceptible to judicial review for abuse of discretion.  The court’s recent treatment of the dispute over Governor Dayton’s veto of legislative funding illustrates the court’s instincts (correctly, in my view) to avoid deciding these interbranch conflicts whenever possible. Ninetieth Minnesota State Senate v. Dayton, 903 N.W.2d 609 (Minn. 2017). Minnesota, however, does not really have an established political question doctrine. But the desire to keep the court out of conflicts of this nature might incline it to hold the constitution confers unreviewable discretion to the governor.

Third, assume that section 12.31’s deeming of all longer duration peacetime emergency declarations to be “extraordinary occasions” is constitutional, then, can the statute also compel exercising gubernatorial discretion to call a special session? The constitutional language provides little insight on this issue. Unlike some state constitutions, it gives the legislature no role in calling special sessions. Nor does it explicitly authorize (e.g., by saying “as prescribed by law”) legal restraints on or expansions of the governor’s power. That silence, I suppose, could be read to imply unfettered discretion that cannot be constrained by statute. But that seems like a weak inference to me.

A modest argument for the statute’s validity might be that the decision to call a session (apart from deciding whether the occasion is “extraordinary”) seems somewhat more legislative. Put another way, deciding whether legislative action is needed or not seems inherently or predominantly legislative in nature (what else could it be?). That might argue against an inference that governors were intended to have unlimited discretion in making such a decision with legislative elements and that it is therefore permissible for a statute to specify in extreme circumstances that the governor must call a session. Moreover, the argument might go, something that rises to the level of peacetime emergency requiring multiple 30-day periods (as here) is just that.

Another factor may argue against the statute’s validity: The context here, which may matter, is that the statute confers executive power on the governor and, then, attempts to put an ad hoc or institutional limit on that power (i.e., by ensuring the legislature is in session so it can terminate the declaration using an arguably unconstitutional procedure). The normal method would be to impose specific directions on when and how the power may be exercised, to provide conditions, which when met terminate the declaration, or to simply set a maximum time limit on the duration. Here, the perception may be that the legislature, through the statute, is attempting to de facto control execution of the law, rather than legislating, its constitutionally assigned duty. Creating a questionable legislative veto procedure (as the statute does) may reinforce the court’s perception that the legislature is using the statute to usurp the governor’s constitutional powers – essentially having its cake (giving the governor the necessary executive power to address emergencies, such as public health crises) and eat it too (giving the legislature a quasi-executive role by convening a legislative session – giving the legislature the ability to exercise power, which it does not have when it is permanently out of session). Put more succinctly, the statute arguably is compelling the governor to call a special session not so that the legislature can legislate, but rather so it can interfere with an executive power, that is, to exercise a constitutionally-suspect legislative veto of the gubernatorial declaration.

If the statute’s special session requirement is unconstitutional, then severability questions arise – does that invalidate the entire statute (unlikely) or other parts of it such as the ability to extend a duration beyond 30 days? The court would need to be convinced, notwithstanding a general presumption for severability, that the intent was to prevent any extension beyond 30 days, for example.  Section 654.20 sets out a high bar for finding provisions are not severable (“so essentially and inseparably connected with, and so dependent upon the void provisions that the court cannot presume the legislature would have enacted the remaining valid provisions without the void one”); it seems likely to be severable.

Bottom line: If Governor Walz decides to extend his declaration in June and for tactical or strategic reasons that he would rather not call a special session under section 12.31, subdivision 2, he has a reasonable case for ignoring that requirement on the grounds that it is unconstitutional.  If the matter is litigated, how the courts would resolve it is unclear. If I were forced to wager on an outcome, my bet is it would be found to be an unconstitutional limit on the governor’s discretion to call special sessions.

MMB released its interim May budget forecast and, as widely expected, it showed an uptick in projected spending with a large reduction in state revenues sufficient to yield a shortfall bigger than the budget reserve balance (but not if you add in the cash flow account).

The following statement from the document captures the essence of the massive uncertainty the state faces:

“The shock to the U.S. economy from the pandemic is unprecedented in modern, post-war history, and the economic outlook is exceptionally uncertain and volatile. Economic outcomes will depend critically on the pandemic’s course, the prospects for an effective treatment and vaccine, government restrictions on economic activity, consumers’ and businesses’ responses as those restrictions are lifted, and the impact of fiscal and monetary policies.” (p. 2)

As if that weren’t bad enough, the Council of Economic Advisors threw cold water on the IHS (the state’s macro forecasting firm) forecast: “First, most Council members believe that a slower recovery is more likely than IHS’ fairly rapid return to economic growth. . . Second, some Council members note that IHS expects both corporate profits and the stock market to reach and exceed pre-crisis levels next year. Those Council members think a less aggressive recovery path for these variables is more likely.”

MMB adjusted it forecast variable as a result of these concerns. Quite frankly (as recognized by the forecast document), we have not experienced anything like this, so it is hard to know how individuals and firms will behave in response, much less the course the virus will take and how effective the public health and medical communities’ responses will be. Truly an awful situation and impossible to forecast with any certainty.

Some reactions:

• An important factor is the uncertainty surrounding how the federal government will respond to the needs of state and local government. To me, this is the elephant in the room. How will the Trump, McConnell, and Pelosi negotiations turn out relative to providing aid to states and locals? This isTHE factor that determines how quickly state budget adjustments should be made. I would expect it to be resolved sooner rather than later, but who knows? McConnell’s bankruptcy statement (since walked back) may seem like nothing more than a negotiating ploy to extract concessions from the Democrats. But given the current GOP and the fact that COVID-19 is perceived as disproportionately a blue state problem, that may be wishful thinking. Trump and congressional Republicans may perceive this is a chance to double down on their motivation for the $10K SALT deduction cap.
• The projected reduction in corporate franchise tax revenue seems too low to me, if only relative to the estimated drops for the income and sales taxes. I will be surprised if corporate revenues do not drop by much more than 13%. That’s based on nothing more than intuition and the fact that those revenue typically drop a lot in any recession. This one may be different.
• On a nitpicky level, I wonder if they included the effect of the CARES Act’s RMD holiday, which will automatically flow through to individual income tax revenues. MMB usually picks up those effects but I didn’t see a reference in the forecast document, which would be typically noted in a regular forecast writeup (this one is a lot shorter, though). The effect is small, so it’s not a big deal either way.
• Overall I expected things to be worse. (Media reports on the California forecast are closer to what I expected.)
• My fear is that this will be like the early 1980s (early in my career) when each quarter a new forecast was done and seemingly (my memory is probably exaggerating) showed a new deficit. That just as to the pattern – on an absolute basis, I expect the downturn to be much worse than the 1980s double-dip recession. A natural reaction by many individuals will to be increase savings for economic reasons and to eschew much consumption because of health concerns, as well as not having as much income to spend. Conversely, firms would seem to have little incentive to make investments, the cost of which they perceive they may never recover. Unlike the 1980s, real interest rates are zero or lower. So, cutting interest rates won’t unleash investment and consumption unlike in the early 1980s. I expect a very slow recovery, but I’m a Cassandra, typically, and have no macro-economics expertise (zero).

Some thoughts on what the governor and legislature should do:

• I’ve never been a fan of kicking the can down the road, but this probably a time to do that (to a limited extent) until we have a better fix on what Washington is going to do about aid to states and local governments. Only the federal government has the legal and financial ability to effectively borrow to maintain government services. One hopes and expects that Congress will recognize that responsibility as it has in the past. You can’t, however, wait too long. If the feds do not come through, it is much better to begin making budget adjustments ASAP.
• I would give MMB expanded unallotment power (e.g., up to each dollar drawn out of the budget reserve or something like that) providing more flexibility to cut spending without fully draining the budget reserve. Cutting sooner, rather than later, will allow more gradual cuts.
• If the feds do not allocate a generous amount of aid, everything will need to be on the table. In days of yore, the time honored method was 1/3 cuts – 1/3 tax increases -1/3 shifts (payment deferrals or revenue accelerations). The reality that Democrats will need to recognize is that 21st century Republicans (at least in a purple state like Minnesota) are unlikely to ever agree to a tax increase. I think it is fair to say that would cast them as RINOs. [My view of the politics: In a deep red state, Republicans will bow to the inevitable reality that taxes are necessary to fund basic government services and will vote for tax increases – gas tax and sales tax increases typically – when necessary. But that’s an intrafamily matter, where their identity as Republicans is not at stake. In a purple state like Minnesota, the game plan is to force Democrats to be responsible for any and all tax increases and to remain the party defined by tax cutting in any and all situations. The days when Al Quie and Arne Carlson would add some Republican imprimatur to modest tax increases are long gone.]
• If the feds fail to pass meaningful state aid, some possible budget fixes include:
  • Reopening state employee contracts as Senator Gazelka has suggested. That will minimize layoffs as compared to straight cuts. It’s true, as some elected Democrats have stated, that cutting public employee compensation won’t “help” the economy. But that is wrong frame of reference if the alternative is lay-offs and cutting public services, in my view.
  • Rescind parts of the 2019 tax cuts. Clearly, the legislature would not have enacted them if it had presciently known the current state of the budget. If the state is going to cut back spending authorized by the 2019 legislature, some of the tax cuts (e.g., the individual rate cut and the working family credit increases) should be undone as well.
  • Similarly, if the legislature and governor are going to review direct state spending to find low priority items or programs that are performing poorly to cut, efforts should not stop there. The host of state tax expenditures (e.g., the surfeit enacted in the 2017 and 2019 tax bills, but I would not limited it to only those) should be evaluated for effectiveness and many likely repealed as wasted efforts to change people’s behavior that either don’t work or aren’t worth whatever small effect they have. Direct and tax expenditures should be treated equally. Protecting poorly designed or low priority tax expenditures because repealing or cutting them can be characterized as a “tax increase” is simply stupid, in my view. But I’m not a politico, so this is probably little more than wishful, wonk thinking.
  • Reinstating some of the state school aid “shift” (i.e., increasing the fiscal year aid hold back). This is bad policy but it is one way the state can indirectly borrow to finance a deficit – i.e., by compelling schools to do so by delaying their state aid payments. If the feds won’t borrow to help, the state will practically need to do so to buy time for budget adjustments. Appropriation bonds are the other option (probably a worse option because the state’s use of deficit financing will be more transparent to Wall Street, dinging the state’s credit reputation).

Of course, no such canon of statutory construction exists, except as imagined by law professors (e.g., in this Harvard Journal of Legislation article by Jesse M. Cross, a University of South Carolina law professor and former congressional drafter).

As a long-time legislative drafter who made more mistakes than I care to admit, I would have welcomed such a canon. It might have helped me sleep better. Unfortunately, it’s only a professor’s pipe dream – one no textualist judge will likely ever buy. I suppose one could codify it (e.g., in chapter 645). It might be interesting to see if one could craft bill language to do that, but I won’t try. That is not what Cross proposes, but rather that courts adopt it as an expansion of an old, infrequently used canon, the scrivener’s error doctrine – a canon used to correct obvious transcription or typographical errors.

It’s worth reading the article if you are interested in the legislative process, statutory construction rules, or (my prime reason) details about how members of Congress actually find out what is in the bills they vote on.

The article has three parts:

• Cross’s description of how Congress has come to heavily rely (exclusively, for all intents and purposes) on staff to draft bills and how members rarely read them – even bits and pieces of them (no surprise to someone who has worked in Minnesota legislature – one assumes Congress is a more extreme version of state legislatures);
• Discussion of the Supreme Court’s “fixing” of a drafting error in the ACA in King v. Burwell, rather than bringing down much of the law’s superstructure by slavishly following a simple “plain language” reading of the provision – this is his “hook” for expanding the scrivener’s canon; and
• Cross’s proposal for the new canon and how he tries to justify it as consistent with statutory construction rubrics, including textualism.

I found the most interesting part to be his description of how Congress has done its drafting – from members drafting in the 19^th century, relying on executive branch personnel in the mid-20^th century, and ultimately on staff by the 1970s – and how members get information about what’s in the bills they vote on.  Some interesting tidbits:

• He presents details on the explosion in offices and staff numbers in the 1970s, when CRS, CBO, etc. were created, as well as the doubling and tripling central agency, committee, and member staff.
• Cross conducted (or had others conduct) interviews of 75 or so congressional staffers as to how members (and higher level staffers such a chiefs of staff and similar) learn about bill text, confirming what we all know: members rarely read bills, relying instead on short documents and oral briefings. These documents are typically staff distillations of more detailed summaries prepared by other staff of the actual bill language. The same pattern generally applies in the Minnesota legislature, based on my experience. Very few members (“outliers” is how Cross describes them) read bill language. Most members understand bills only at the conceptual or purpose level.
• Members of Congress often have 2-page (or shorter) limits on the length of these documents. (Some Minnesota legislators have similar impulses – I remember one Speaker who had a strict 1-page rule and my efforts to pare and condense, as well as to narrow margins, decrease font size etc to bend the rule.)  Many members of Congress apparently use “vote rec” memos (short written recommendations on how they should vote on the floor) (p. 105). They can be as short as index cards that are handed to the member shortly before floor votes.
• As Cross puts it, members’ understanding of bills is almost always “conceptual not text based” (p. 108).
• Cross makes the point that this limited engagement with bill text is not just because members do not have time or are too busy (making campaign finance fund raising calls?) to read bills. Rather, the nature of federal legislation makes it impossible for generalist members to consider and deliberate over bill text – (1) the world/problems being addressed are too complex, (2) the statutes themselves are too specialized and labyrinth, and (3) agency administration of the programs adds another layer of complexity.  The system expects members to be generalists who simply cannot have the specialized expertise necessary to understand the details of what they’re voting on (i.e., to be held responsible for the staff mistakes, apparently), even if they had or could make time to do so.
• Cross reports that the Senate and House bill drafting offices consider that it takes between 2 and 6 years to train drafting staff in subject areas (pp. 113 -114).  That’s longer than in the Minnesota legislature, where I would guess the expectation is closer to 1 to 2 years.  It may be a little longer in specialized areas.
• His bottom line: By necessity, this has resulted in a legislative bureaucracy to which members have delegated responsibility for drafting and making sure that the text of bills is consistent with members’ purpose in passing legislation. Of course, much of detail of policy development has also been delegated, since making many smaller policy choices requires descending into the details of the problem, the statute, and administrative practices that members have not mastered.

This delegation or “division of labor,” as Cross puts it, is crucial to his theory for expanding the scrivener’s error doctrine. Drafters are now the scribes from days of yore (Scalia’s term, not Cross’s) whose obvious mistakes can be corrected by the courts. Aside from that similarity to mistakes made by scribes, he thinks it is inapt to hold elected members responsible for mistakes made by their agents (staff). In his view, doing so would undercut democratic theory. I am not sure that I agree – Harry Truman’s the “buck stops here” and all that.  It’s not clear why a legislature should be different. More on that is below under my reaction.

The second and third part of the article were less interesting to me, so I won’t spend much time describing them.  Cross argues that the Court’s resolution of the dispute in King v. Burwell could be used as a template for generalizing the scrivener’s error doctrine into a broader staffer’s error doctrine to correct drafting mistakes (“isolated snippets of statutory text” as he puts it) that are manifestly contrary to the purpose of the legislation.  He contends that doing so would be consistent with either of the two schools of statutory interpretation – intentionalism or textualism.

The Burwell hook. To grossly oversimplify, in King v. Burwell the Court was faced with a drafting error in the ACA that would have denied the federal subsidy for those purchasing coverage through the healthcare.gov website (i.e., in states that did not set up their own exchanges), if the Court followed a simplistic reading of the language of the law.  (It could have resolved the matter by deferring to the agency interpretation under Chevron principles, but it did not want to do that – a different story.) That reading would have denied the subsidies to residents of many states, undercutting a central feature of the ACC: clearly not what anyone thought they were voting on.  Apparently, that was too much for Chief Justice Roberts (but not for Justice Scalia who dissented joined by Thomas and Alito), because it was so clearly contrary to what Congress intended. Thus, he deviated from the “natural reading of the pertinent statutory phrase” as he put it and instead read it (“unnaturally”?) to avoid thwarting Congress’s intent.

Expanding scrivener’s error doctrine. Cross runs with the facts and resolution of Burwell to formulate a doctrine that addresses similar situations by expanding the scrivener’s error doctrine to a broader class of staff drafting errors that are clearly inconsistent with a statute’s purposes.  To apply Cross’s canon, a court would go through three steps, to:

1. Identify the overarching policy goal of the statute.
2. Determine if an “isolated snippet of statutory text” (i.e., the error) would “sabotage those overarching policy goals.”
3. Make sure it really is error (not just the court’s misperception of the policy or intention).

He goes on to make good arguments for its adoption by intentionalists and plausible arguments for textualists.

My Reaction

As usual, reading an article like this gets my mind to running off on tangents.  The following is a distillation of a few of my off-the-cuff reactions for whatever they may or may not be worth.

Burwell is probably not a good case to use as a foundation for a new statutory construction canon.  I’m skeptical of the prospect for Cross’s idea (much as I might like it) being adopted by federal courts. Burwell is a unique case; a “hard” case that a textualist would likely consider made “bad” law because of its extreme circumstances, probably not something to generalize and go back on textualist principles (more on that below).  Specifically, I have two observations about the decision:

• If a more ordinary law had been at issue (e.g., a securities regulation or something similar) and the “natural” reading of the language would have benefited business, I doubt Roberts would have hesitated about going with the “natural reading” no matter how inconsistent it was with an overarching and obvious purpose of the law.
• I would guess it was only the highly politicized situation in Burwell that made him unwilling to go there: doing so would have vitiated the signature policy of the Obama administration (and a goal that generations of Democrats identified as one of their highest priorities) and would have created a perception by many that the Court had essentially played a game of gotcha with the president and congress.

Textualists will not be impressed or sympathetic, I think. Reading the article reminded me (as an intentionalist) how much I dislike textualism, especially in its more rigid versions (mainly because I think relying on dictionaries and grammar rules divorced from consideration of purpose and extrinsic evidence enables judges to do more mischief than text anchored by a beacon of legislative purpose).  I cynically have thought that the infatuation with textualism (over the last 25 years or so) was driven by largely an anti-government, “conservative” ideology (gum-up government to express it even more cynically). If that is at all accurate, that underpinning will be inhospitable to a canon whose prime attraction is help the wheels of government keeping moving as intended, notwithstanding unintended staff drafting errors. I doubt Cross’s canon would appeal to committed textualists, like Scalia’s textualist acolytes.

I have suspected that there is more of a political and philosophical agenda at stake in rigid textualism.  To put a slightly finer point on it, my perception is that the textualist movement is heavily driven by two elements:

• Distrust of liberal or expansive-government supporting judges who, by resorting to effectuating “legislative intent” as evidenced by legislative history or their imaginations of what was intended, recast/rescue/remold statutes to achieve their ends (more government). Requiring strict adherence to the dictionary meanings and rules of grammar (obvious inconsistent purpose be damned) limits that discretion and prevents them from doing mischief.  This can be cast as supporting democratic values, because (federal) judges are not elected and judges (elected or not) are not supposed to “make policy.” In reality, applying “plain language” rules (dictionary and grammar meanings divorced from intent and extrinsic evidence of what the legislature intended) are remarkably malleable and can allow surprisingly unintended interpretations in the hands of ideologically motivated judges.
• A desire to hem in or limit legislation – a preference for thwarting legislation rather than effectuating or helping to faithfully implement its purposes.  To me it seems almost inevitable that slavishly following the text will frequently make it more difficult to effectively legislate (a good result for someone who favors small and constrained government). This is so because it will require more runs through a legislative process that favors playing defense (preventing change). One might assume that strictly binding courts to the rigid text would be equally probable to break either way – i.e., in favor or against bigger government.  I think that is unlikely to be the case, because impractical and egregious mistakes that disadvantage private actions are often ignored by the executive on grounds of impracticality. It’s more than ties go to those favoring limited government – short of drafting very general provisions (and investing a lot discretion to the executive, which may not be favored for a host of reasons), it is very hard (impossible) to anticipate the multiplicity of fact situations and future developments.  Thus, I see the attraction of rigid textualism to those who favor more limited or less effective government.  As a result, it undercuts the power of the legislative branch by raising hurdles that must be cleared by staff drafters – getting the text exactly right (anticipating the unforeseeable) as well as avoiding mistakes.

So much for my ranting.

How much would it matter? Assessing that is key (for a judge considering adopting the canon or for a legislature thinking of codifying it). The key point would be the fuzzy step 3 (and to a lesser extent step 2), I think, and how that is expressed and applied. I don’t have a good feel for that. I think the current system (before the advent of rigid textualist) coped pretty well.

Would staff evidence of their mistakes be considered? Focusing on Burwell may make it easy to overlook that there will be inevitable questions of whether or not provision really was drafted in error. (Of course, for Cross that’s not enough – a mistake needs to thwart an overarching purpose.) Since I was many times asked to attest to legislative intent, one naturally assumes that staff would be pressed to admit/testify to having made a mistake contrary to what the author/committee/etc wanted. That raises some interesting questions. I assume that evidence for the existence of the mistake and its contravening an overarching purpose would need to be contemporaneous with passage, so it could not be ginned up or augmented after the fact. Cross does not mention any of that as far as I can remember.

Division of labor and should it matter? As noted, Professor Cross relies heavily on the idea that it is not appropriate to hold generalist legislators responsible for the drafting errors of their staff agents. As a former drafter, that sure resonates with me personally. But the actual logic is less clear to me. Would he apply the same rule to regulations that have the force of law and go through an executive branch approval process? The same division of labor probably exists there (drafter versus higher up that ultimately must sign off)? Because of the practice of explicitly stating purposes of regs and administrative rules (typically anyway), a clearer understanding of purpose would seem to be present to test against. Entity based intent also seems less of a synthetic concept for regulations than legislation, making it an easier choice to invalidate an inconsistent drafting error. Would the result be different in a parliamentary system (e.g., the UK) where the executive and legislative branches are unitary and legislation could be considered a product akin to an administrative regulation? Is the problem that in America each legislator is considered more a free agent and that drafting and review is typically done by central agency staff who are not directly responsible to or supervised by the members (Cross doesn’t discuss this element of the process)? Lots of interesting questions to me that the article does not plum.

Polarization makes this a bigger problem. Anyone involved in the legislative process knows that drafting errors regularly occur. Everyone makes mistakes and some of them slip through even the best of quality control systems. Moreover, as the old proverb says, “haste makes waste” and the legislative process is characterized by much haste. So, no surprise, drafting errors are not an uncommon occurrence – if only because there is not enough time for several layers of carefully checking, even if that is what operational protocols call for.

A recent WaPo article provides what appears to be a good example of a recent mistake in a relatively high profile federal law, the December budget deal, which granted parental leave benefits to federal employees.  According to the article, the benefits were not extended to a fair number of employees as a result of what appears to be a cross reference error (i.e., not listing all the relevant statutes to cover the intended employees). That is a classic type of error that can easily occur.

Traditionally, this sort of thing was not a serious problem with a collegial legislature that recognizes errors occur and routinely corrects them, even if doing so cedes a modest advantage to the other side. In the federal tax legislative process that was regularly done by enacting technical correction legislation. In Minnesota, similar technical correction fixes were also routinely done in omnibus tax bills.  Both of those practices occur less commonly now, if at all, unfortunately.  In the past, most members recognized that fixing unintended (staff) mistakes was the right thing to do to honor legislative intent and/or agreements on deals (even when there was change in control after the bill’s enactment).

Polarization and a “take no prisoners” warfare approach to the legislative process has made it increasingly unlikely that errors are corrected in controversial legislation. Elections that empower an “out” party to block legislative fixes by taking control of one house or the executive are a factor too. The ACA is, of course, the high-profile example. For the ACA the depth of the Republican opposition caused years of litigation that attempted to use a drafting error/oversight to invalidate a core part of the legislation. Two of the Supreme Court conservative majority stepped back from a talismanic commitment to plain language and textualism as an excuse to bring down a generational change in how America provides health benefits to a large segment of its population.  I doubt that can be generalized to less high profile example, much as it might be good idea.

Possibility of state adoption. Cross’s article focuses on how congress drafts and reviews bill text. As a result, it has strictly no application for state courts interpreting state statutes. I suspect that his arguments would apply almost as well to larger and more sophisticated state legislatures (certainly the New Yorks and Californias). And it is courts in some of those states who are more likely to be open to his ideas, unfortunately. Just a thought.

Bottom line: the article provides some good background information about the Congressional processes and folkways, as well as raising some interesting questions, even if its proposal is unlikely to be adopted.