Inspired by the posting, I have revised my Federal Tax Procedure text to include the points that some Tax Court appeals are exclusively to the Court of Appeals for the District of Columbia Circuit and that, at least as to those appeals for now, the application of the Golsen rule will mean that the District of Columbia Court of Appeals will set uniform national law on the issues, meaning that certiorari to the Supreme Court will be based exclusively on importance of the issue and not on conflict among the circuits. (That is overbroad, but a sufficient generalization for now.)
The concept of a single court of appeals, variously formulated, for tax cases has been around for many years, with many proponents and opponents. It is a long history which is suggested by the following quote from an article (in a footnote, no less): Ruth Bader Ginsburg and Peter W. Huber, The Intercircuit Committee, 100 Harv. L. Rev. 1417, 1429 n. 61 (1987):
There may be a few discrete bodies of law so arcane and complex that no other solution will do. The Federal Circuit now satisfies the need for early appellate declaration of national law in certain areas, notably, patent disputes. See 28 U.S.C. § 1295 (1982) (assigning to the United States Court of Appeals for the Federal Circuit exclusive jurisdiction over enumerated matters, including appeals from the district courts in patent cases, appeals from the Merit Systems Protection Board, appeals from the agency boards of contract appeals, and appeals from the district courts in certain cases against the United States); see also S. REP. NO. 275, 97th Cong., 2d Sess. 3, 4, reprinted in 1982 U.S. CODE CONG. & ADMIN. NEWS 11, 13, 14. Congress believed that the federal judicial system lacked sufficient capacity "to provide reasonably quick and definitive answers to legal questions of nationwide significance." Id. at 13. It therefore established the Federal Circuit to adjudicate definitively in areas where the legislators found "special need for nationwide uniformity." Id. at 14.
A single court of tax appeals could promote uniformity and coherence in another federal law domain populated by specialist advocates and rarely benefited by the labors of generalist judges, including those on the Supreme Court. See H. FRIENDLY, FEDERAL JURISDICTION: A GENERAL VIEW 161-71 (1973); Ginsburg, Making Tax Law Through the Judicial Process, 70 A.B.A. J. 74 (1984); Griswold, The Need for a Court of Tax Appeals, 57 HARV. L. REV. 1153 (1944).In the course of the research I picked up that two leading legal authorities, Roger Traynor and Stanley Surrey had also proposed such a single court of tax appeals. See Mirit Eyal-Cohen, Preventive Tax Policy: Chief Justice Roger J. Traynor's Tax Philosophy, 59 Hastings L.J. 877 (2008).
For fans of tax procedure, I cite a work of Surrey and Traynor and two others as the best tax procedure law review article ever written. John M. Maguire, Stanley S. Surrey and Roger John Traynor, Section 820 of the Revenue Act of 1938, 48 Yale L. J. 509 (Part 1) and 719 (Part 2) (1939). This article discusses the enactment and scope of the original mitigation provisions in Sections 1311-1314. I love the mitigation provisions, which may be why I favor the cited article. My favorite quote about the mitigation provisions is from then Solicitor General Erwin Griswold (former dean of the Harvard Law School and prominent tax proferssor): "We can't take a mitigation case to the Supreme Court; it [the Supreme Court] will never understand mitigation." [That may not be an exact quote, but it is close and certainly the sense is right.]