Lochner’s legacy: the real reason we can’t have a graduated income tax in Massachusetts

The conventional wisdom for years in Massachusetts has been that, in 1915, the people adopted Amendment Article 44 of the state Constitution which bars a graduated income tax.  Therefore, if you want to implement such a tax, you have to amend the state Constitution.

I’m pretty sure that’s not the whole story.  What I think actually happened is that, in 1915, the people adopted Article 44 as a way of clamping down on tax avoidance schemes that were crippling the Commonwealth’s ability to enforce its tax laws and hence costing it a lot of money.  Unfortunately, some of Article 44’s wording is on the opaque side, so inevitably it ended up in the Supreme Judicial Court.  When it did, in 1921, the SJC relied on the Lochner era’s then-popular but now discredited understanding of the nature of contracts and property to rule that earned income was “income derived from property.”  And, because Article 44 requires that the tax rates on income derived from property must be uniform, it follows that we can’t have a graduated income tax.

In my view, this result was intended neither by the people who drafted Article 44, nor, most likely, by the people who voted for it at the ballot box.  Article 44, correctly understood, simply says nothing at all about a graduated income tax – doesn’t require it, but surely doesn’t bar it either.  So it wasn’t the people who drafted or voted for Article 44 who prevented us from having a graduated income tax.  Rather, it was five misguided Justices of the Supreme Judicial Court.  The constitutional doctrine on which they relied, which was the centerpiece of the period now known as the Lochner era, has long since been relegated to the dustbin of history.  But its detritus remains in the form of a judicial mistake that to this day hampers our ability to have a progressive tax structure in Massachusetts.

Details on the flip. #longreads

Let’s start, as always with inquiries like this one, with the text of the provision at issue.  Here it is, in full.

Full power and authority are hereby given and granted to the general court to impose and levy a tax on income in the manner hereinafter provided. Such tax may be at different rates upon income derived from different classes of property, but shall be levied at a uniform rate throughout the commonwealth upon incomes derived from the same class of property. The general court may tax income not derived from property at a lower rate than income derived from property, and may grant reasonable exemptions and abatements. Any class of property the income from which is taxed under the provisions of this article may be exempted from the imposition and levying of proportional and reasonable assessments, rates and taxes as at present authorized by the constitution. This article shall not be construed to limit the power of the general court to impose and levy reasonable duties and excises.

The key sentences are the second and third, which talk about income “derived from property” and “not derived from property.”  Here’s the crux of the matter: if earned income (by which we normally mean salaries and wages) is “income derived from property,” then it must be taxed at a uniform rate.  But if it’s not, then Article 44 places no restriction on how it may be taxed.

So, is a salary “income derived from property”?  Not in what I would take to be the ordinary meaning of those words.  “Income derived from property” seems to me most naturally to apply to, for example, rents obtained from real estate, or dividends obtained from stocks, or gains realized by selling something whose value has increased over time, or other means by which one generates cash from something that one owns.  But one has to get pretty esoteric to stretch “income derived from property” to include a salary or a wage.

Unfortunately, that is exactly what the SJC did in 1921.  In a case called Raymer v. Tax Commissioner, helpfully called to my attention by BMGer Hester Prynne, a Harvard professor, one George Raymer, complained that his salary from Harvard was being taxed at a higher rate than income derived from annuities – unquestionably “income derived from property.”  Raymer argued that his salary was not “income derived from property,” and therefore that under the language of Article 44 (which says that “the general court may tax income not derived from property at a lower rate than income derived from property”), the tax on his salary was unconstitutional.

The SJC rejected Raymer’s argument by concluding that Raymer’s salary was “income derived from property.”  What sort of “property” might that be, you might well ask?  Here’s the SJC’s reasoning:

“Property” is a word of large import. It has been interpreted as including the right to make contracts for labor and for personal service. Decisions to that effect had been rendered by courts of the highest authority before the adoption of the Forty-fourth Amendment. It was said in Coppage v. Kansas, 236 U. S. 1, at page 14 (affirming earlier decisions in this particular), “Included in the right of personal liberty and the right of private property partaking of the nature of each – is the right to make contracts for the acquisition of property. Chief among such contracts is that of personal employment, by which labor and other services are exchanged for money or other forms of property.” This was but a redeclaration of the principle of Adair v. United States, 208 U. S. 161. Cases to this point are collected and reviewed, in Bogni v. Perotti, 224 Mass. 152, 154, 155, where the same conclusion was reached. It is not open to question that contracts for labor and service are “property” within the meaning of that word in both the Federal and State Constitutions, and as such they are entitled to the protection of the numerous guarantees thereby afforded. It would be a strained and unnatural construction to hold that that which was “property” for the purposes of the protections afforded by the Constitution was not “property” for the purposes of the taxation of income “derived from property” authorized by an amendment to this same instrument.

Got that?  What the SJC is saying here is that a contract by which one agrees to perform some sort of labor in exchange for being paid is “property,” and therefore that the income you derive from that contract is “income derived from property.”

The lawyers and legal historians out there have probably just heard a loud bell going off in their heads.  Why?  Because the notion that a contract for labor or service in exchange for money is a form of “property” is a key feature of the Lochner line of U.S. Supreme Court cases.  That notorious line of cases held that, because contracts of that kind were “property,” state or federal regulations designed to protect workers – e.g., that limited the number of hours a baker could work, or that barred railroads or other employers from firing workers for joining a union – were unconstitutional because they impinged on both the employers’ and the employees’ constitutional right to enter into whatever contract they wished.  Thus, they violated the 14th Amendment’s guarantee that the government may not deprive individuals of life, liberty, or property without due process of law.

The SJC’s Raymer opinion is singularly unconvincing for a couple of reasons.  First, the Lochner line of cases was repudiated by the Supreme Court in 1934, and today nobody takes seriously the notion that there is a constitutional right to enter into contracts free of any state oversight.  One can’t expect the SJC to have anticipated a doctrinal shift that was at that point 20 years in the future, but at the same time, one might have expected the SJC to think for itself rather than adopt wholesale an unconvincing line of cases that were decided in a context quite different than that presented in Raymer.

Second, and perhaps more damningly, the SJC’s adoption of Lochner-style views on contracts rendered an important clause of Article 44 almost entirely meaningless.  The third sentence of Article 44 says that “[t]he general court may tax income not derived from property at a lower rate than income derived from property.”  Thus, the drafters of Article 44 thought that “income not derived from property” was an important enough species of “income” that it warranted special constitutional treatment.

But under the rule of Raymer, what sort of income would qualify as “income not derived from property”?  If every contract designed to generate some sort of income is “property” – and that’s basically what Raymer says – one is hard-pressed to come up with anything at all that would qualify.  Our resident tax whiz, power wheels, suggested that inter vivos gifts or money found in the street would qualify, and that’s fair enough.  But it’s hard to imagine that those trivial sources of income were deemed sufficiently important to receive a special constitutional provision.  Rather, one imagines that the drafters of Article 44 had something significant in mind when carving out a special rule for “income not derived from property,” and the most obvious candidate would be earned income.  It’s a cardinal rule of judging that one is supposed to avoid interpretations that render portions of the text being interpreted meaningless, yet that is precisely what the Raymer Court did.

The SJC itself has acknowledged this problem.  In a footnote to a 1981 Opinion of the Justices, the Justices observed as follows (emphasis mine):

The third sentence of art. 44 provides that the General Court may tax “income not derived from property” at a lower rate than “income derived from property.” This may have been intended to allow for a lower rate of taxation for earned income. See P. Nichols, Taxation in Massachusetts 471 (3d ed. 1938). But in Raymer v. Tax Comm’r, 239 Mass. 410, 413 (1921), the court held that the salary of a university professor was “income derived from property” within the meaning of art. 44, making the significance of the third sentence of the amendment unclear. What is important for our purposes here, however, is that it is generally agreed that the requirement of uniform rates of taxation in art. 44 applies to both earned and unearned income.

The Justices in that opinion had evidently not been asked, and thus were not prepared, to reconsider Raymer‘s interpretation of Article 44.  But I find it interesting that they went out of their way to point out its obvious deficiencies.

From all of this, it is clear to me that the Raymer Court’s interpretation of Article 44 was an unfortunate by-product of a misguided and now-discarded legal doctrine, and that earned income should never have been deemed to be “income derived from property.”  But what about the uniformity requirement for “income derived from property” – is there a good explanation for that, absent any consideration about a graduated income tax?

Yes there is – a very good explanation, actually.  I talked about this a bit in my last post on this general topic.  To summarize briefly, the original state Constitution conferred upon the state and municipalities the power to tax, as long as the taxes were “proportional.”  To again quote from Justice Cordy, who explored this question in a 2004 opinion, this rule

was repeatedly construed to require that municipalities (the principal collectors of taxes levied at the time) must apply a single rate of taxation to the total value of the property of all taxpayers in their respective communities, thereby ensuring that each taxpayer paid a share of the municipal burden proportional to his share of the wealth in the municipality.  While each municipality could apply a different rate of taxation on property in its community, the rate applied was required to be uniform for all property throughout that municipality.

The problems began when wealth, which previously had generally taken fairly tangible and thus easily-assessed forms (such as land), began increasingly to assume intangible, easily-moved and -concealed forms (such as stocks and bonds).  That development made it much easier for wealthy Massachusetts residents to move their assets from municipalities with higher tax rates and more efficient tax collectors (generally the large cities like Boston) to those with lower rates and less effective collectors, or to find ways of shielding them from municipal collectors all together.  It seems to have been an early 20th-century equivalent of sheltering your assets in the Cayman Islands.  According to a 1938 book on the subject (as quoted by Justice Cordy), this practice led to “hardly a fifth of the personal property in the commonwealth” being properly taxed.

So along comes Article 44.  And it solves this problem in one fell swoop: by (a) explicitly conferring on the state legislature the power to tax income of any sort, and (b) requiring that income derived from any particular class of property must be taxed at the same rate “throughout the commonwealth,” the drafters seem clearly to have been intending to end the practice of moving assets (“property”) around the state to obtain more favorable tax treatment.  The use of the phrase “throughout the commonwealth,” which seems mainly to be geographical, supports the interpretation that Article 44 was, essentially, a response to the increasingly creative and agressive use within Massachusetts of what we would now call tax shelters.

Now, what about the fact that the drafters of Article 44 apparently considered and rejected a version that would have expressly adopted some sort of graduated income tax?  I agree that that’s interesting, and I would like to know more about it.  In particular, I would like to know whether the rejected proposal involved graduated rates on earned income, or “income not derived from property,” or whether it focused on “income derived from property.”  If the focus was on “income derived from property” (and frankly I suspect that’s the case), then it’s simply not relevant to a discussion of whether Article 44 says anything about a graduated income tax on earned income.  And even if not, we still are faced with the fact – and I believe it is a fact, the SJC’s unfortunate mistake notwithstanding – that Article 44 simply does not say anything about a graduated income tax on earned income.  It doesn’t impose it, but it doesn’t bar it either.  It seems to me entirely reasonable to conclude that the drafters chose to leave to subsequent legislatures the decision whether or not to adopt such a tax structure.

Law geek that I am, I cannot resist making one further observation: Oliver Wendell Holmes is generally acknowledged to be one of the greatest Justices ever to sit on the Supreme Court of the United States.  Before he got that job, he was Chief Justice of the Supreme Judicial Court of Massachusetts.  Holmes dissented vigorously from the Lochner line of cases, including from the Coppage and Adair decisions upon which the Raymer Court unfortunately relied.  So the folks sitting on the Court that Holmes used to lead relied on cases that Holmes himself had rejected.  And in one of those cases (Coppage), Holmes looked back home, citing a couple of dissenting opinions he had authored on the SJC, and stating that “I still entertain the opinions expressed by me in Massachusetts.”  Sometimes, when you’re right, you’re right, even though nobody agrees with you at the time.  Good for Holmes for sticking to his guns for all those years.

So, where does all of this leave us?  Is it worth asking the SJC to reconsider Raymer?  Maybe … but it also must be acknowledged that Raymer has been the settled interpretation of Article 44 for a long time, and that in a couple of instances the people have been asked to change it via a ballot question, and they have declined the invitation.  Ninety years later, it does not seem unreasonable to conclude that, even if the original interpretation was wrong, it’s now generally understood, and it should be the people rather than the judiciary that makes a change.

Still, I think it’s worth understanding that the SJC’s original interpretation of Article 44 really was incorrect.  It was a result of the SJC’s unfortunate decision to apply an incredibly activist and anti-progressive line of Supreme Court cases, that have long since been abandoned, in a context where they simply made no sense.  Anyone who wants to argue that a flat income tax is good policy is free to do so, but in my view, they must do so without the history of Article 44’s genesis on their side.