This case involves the validity of a federal tax lien, and a jurisdictional and a procedural question. The case grows out of assessments that the Internal Revenue Service made in April 1979 against Unique Industries, Inc., for unpaid taxes. In July the Service filed a federal tax lien with the Recorder of Deeds of DuPage County, Illinois (and also, a few days later, in California but arguably in the wrong office--a point of some significance, as will appear). One month earlier Dimmitt & Owens Financial, Inc., a factor, had begun buying accounts receivable from Unique, obtaining a security interest which it recorded that month in Illinois by filing a financing statement with the Secretary of State of Illinois, and in September in California by filing a similar statement with the secretary of that state. The federal tax lien (provided it was filed in the right place) came into effect on the forty-fifth day after it was filed, 26 U.S.C. Sec. 6323(c)(2)(A), but Dimmitt & Owens, unaware of the lien, kept on factoring Unique's accounts. The Internal Revenue Service served it with a notice of levy on several hundred thousand dollars worth of accounts receivable that Dimmitt & Owens had collected after the (alleged) effective date of the tax lien. Dimmitt & Owens brought suit against the United States under 26 U.S.C. Sec. 7426 to enjoin the alleged wrongful levy. The government counterclaimed, seeking a judgment foreclosing its lien against the proceeds of the accounts receivable. The plaintiff joined as additional defendants Unique, a subsidiary of Unique, and Unique's president, Dudley Olsen. The claim against Unique and its subsidiary was for default on the factoring agreement; against Olsen it was for fraudulent concealment of the liens. The basis of federal jurisdiction against these additional defendants was diversity of citizenship. None of these defendants appeared, and a default order was entered against all of them, and later a default judgment against Olsen.

The plaintiff did not fare as well against the government. The basis of its claim was that the government had filed the tax lien in the wrong place. The lien must be filed at "the place at which the principal executive office" of the taxpaying corporation is located, 26 U.S.C. Sec. 6323(f)(2), and Dimmitt & Owens contended that Unique's principal executive office had been in California rather than Illinois, and that although the Internal Revenue Service had filed in California too, it had filed in the wrong office there. The district judge held that Illinois was the right place to file, and (since there was no question that, if so, the government had filed in the right office in Illinois) granted the government's motion for summary judgment. 589 F. Supp. 14 (N.D.Ill.1983). The judge invited the government to present a judgment order for his signature. The government failed to appear at the appointed time and the judge entered a default judgment for Dimmitt & Owens. Later however he granted the government's motion under Fed.R.Civ.P. 60(b) to vacate that judgment, and entered a new judgment, dismissing the complaint and ordering Dimmitt & Owens to pay the government some $300,000 in satisfaction of the government's lien. Dimmitt & Owens appeals.

The initial question (characteristically not addressed by either party) is whether the judgment is final within the meaning of 28 U.S.C. Sec. 1291, and hence appealable to us. No judgment was ever entered against two of the three private defendants, and no order was entered under Rule 54(b) of the Federal Rules of Civil Procedure that would have allowed the judgment that the district court had entered in favor of the United States against Dimmitt & Owens to be appealed without waiting for the entire litigation to be concluded. A Rule 54(b) order would have been proper because the judgment against Dimmitt & Owens wound up the entire dispute between two of the parties (the government and Dimmitt & Owens). Walker v. Maccabees Mutual Life Ins. Co., 753 F.2d 599, 601 (7th Cir.1985). And, if the other parties remained in the lawsuit, such an order would have been essential to the appealability of the judgment against Dimmitt & Owens--provided the suit that Dimmitt & Owens had brought was really one lawsuit, and not two (one against the United States, and the other against the private defendants). But it was one suit. Although Dimmitt & Owens could have brought separate lawsuits against the United States on the one hand and the private defendants on the other, the joinder of all the defendants in one suit was proper. See Fed.R.Civ.P. 20(a).

Since no Rule 54(b) order was entered, it becomes important whether any of the private defendants are still parties to this lawsuit. A default order was entered against them, it is true, but a default order against a defendant is not a final judgment that concludes the lawsuit against that defendant, even on the matters covered by the order. 10 Wright, Miller & Kane, Federal Practice and Procedure Sec. 2692, at pp. 465-66 (2d ed. 1983). Only against Olsen was a default judgment entered. A default order is an interim ruling which merely establishes that the defendant is liable to the plaintiff. It does not determine the extent of the liability. A plaintiff in whose favor a default order is entered must still establish how much the defendant owes him, and the defendant must be ordered to pay the amount; and these steps were not taken here against Unique and its subsidiary. An order that establishes liability but leaves damages still to be fixed is a classic example of a nonfinal, nonappealable order. See, e.g., Liberty Mutual Ins. Co. v. Wetzel, 424 U.S. 737, 744, 96 S. Ct. 1202, 1206, 47 L. Ed. 2d 435 (1976); Parks v. Pavkovic, 753 F.2d 1397, 1401 (7th Cir.1985).

However, inquiry at argument disclosed that Unique and its subsidiary are in fact no longer parties to this lawsuit. They are in bankruptcy and any effort by Dimmitt & Owens to recoup the cost of this judgment from them will be pursued if at all in the bankruptcy court. Although they have never been formally dismissed in the district court we conclude that they ceased to be parties there before the judgment for the United States was entered, so nothing of the lawsuit was left for later decision by the district court. But there should have been an order dismissing those parties. We implore the bench and bar of this circuit to tie up all jurisdictional loose ends in the district court and not allow unnecessary, and occasionally fatal, jurisdictional uncertainties to dog the appeal. See, e.g., Wisconsin Knife Works v. National Metal Crafters, 781 F.2d 1280, 1282-83 (7th Cir.1986); Kanzelberger v. Kanzelberger, 782 F.2d 774 (7th Cir.1986); Tenneco Inc. v. Saxony Bar & Tube, Inc., 776 F.2d 1375 (7th Cir.1985).

The principal issues on the merits are whether the district judge was correct in ruling that there was no genuine issue of material fact concerning the location of Unique's "principal executive office," and whether he acted permissibly in vacating the default judgment that he had entered against the government after ruling in the government's favor. On the first issue, Dimmitt & Owens contends that Unique's principal place of business was in California rather than Illinois. It supports this contention with cases construing this term as it appears in 28 U.S.C. Sec. 1332(c), which defines the state of citizenship of a corporation for purposes of the diversity jurisdiction. This procedure, however, uncritically assumes that "principal executive office" in 26 U.S.C. Sec. 6323(f)(2) means the same thing as "principal place of business" in 28 U.S.C. Sec. 1332(c), and it need not.

The statutes have different purposes, and statutory language must always be interpreted in light of statutory purpose. The purpose of section 1332(c) in making a corporation a citizen of the state where it has its principal place of business as well as the state where it is incorporated is to exclude from the federal diversity jurisdiction cases between a citizen of a state and a corporation whose center of gravity is in the same state even though it may be incorporated elsewhere; for such a corporation should be sufficiently "local"--sufficiently identified with the state--to avoid the obloquy that may attach to a "foreign" corporation in litigation with a local resident and that provides the modern rationale of the diversity jurisdiction. The words "principal place of business" are to be construed with this purpose in mind.

The purpose of the corresponding words in section 6323(f)(2) is different. It is to fix the place where a tax lien must be filed in order to be valid. Here the most important thing is to have a simple and definite test so that the government knows where to file its liens and prospective lenders (such as Dimmitt & Owens) know where to look for tax liens against the borrower. There are thousands of counties in the United States. It would be preposterous to have a system under which either the government had to file its liens in every county or prospective lenders had to search for liens in every county. It hardly matters whether the place in which to look for liens is also the corporation's center of gravity, the ascertainment of which might require counting employees and toting up assets rather than just locating corporate headquarters. All that matters for purposes of fixing the location for filing and searching for liens is that the place chosen be the simple, obvious, and natural place to file and look for liens. As the Senate report on the bill that became section 6323(f)(2) explained, "principal executive office" was chosen because "this is the most readily identifiable of all the offices that a business may maintain, appearing, as it does, on the annual reports filed with most States and on similar returns, and avoids the uncertainty of determining which of the many business offices that a taxpayer may maintain is its principal one." See S.Rep. No. 1708, 89th Cong., 2d Sess. 11 (1966), U.S.Code Cong. & Admin.News 1966, pp. 3722, 3732. The term "principal executive office" should therefore be interpreted in light of the objective of making the identification of the place for filing and searching for liens as simple and certain as possible.

Since certainty of jurisdiction is a desideratum too--the parties ought to know definitely what court they belong in, and not face the prospect that their litigation may be set at naught because they made a wrong guess about jurisdiction--this circuit has long used a simple "nerve center" test for principal place of business. See, e.g., Sabo v. Standard Oil Co., 295 F.2d 893 (7th Cir.1961); Kanzelberger v. Kanzelberger, supra, 782 F.2d at 777. In Sabo we used the term "executive offices" as a synonym for nerve center, see 295 F.2d at 894, and that led us in McKee-Berger-Mansueto, Inc. v. Board of Education, 691 F.2d 828, 833 n. 4 (7th Cir.1982), to equate the test for principal place of business in section 1332(c) with that for principal executive office in section 6323(f)(2), although, so far as appears, in neither Sabo nor McKee-Berger-Mansueto was there a suggestion that the real nerve center of the company might be elsewhere than at its executive offices. The result of the footnote in McKee-Berger-Mansueto is that, in this circuit, principal place of business and principal executive office mean the same thing, though adventitiously. Dimmitt & Owens overlooks the essential qualification: the cases it has cited to us on principal place of business are from circuits that reject the nerve center test in favor of a broader test which requires consideration of such things as where the assets and employees of the corporation are concentrated.

Whatever the merits of the broader test in helping to ascertain a corporation's state of citizenship for purposes of diversity jurisdiction, we think it quite out of place in the tax lien context, where certainty plays an even more central role and where the policy of confining diversity jurisdiction to cases where there is a real and not merely theoretical danger of prejudice to an out-of-state firm is irrelevant. For purposes of section 6323(f)(2) it is best to ask simply, where was the obvious place for the Internal Revenue Service to file its liens against Unique and for Dimmitt & Owens to look for possible liens against Unique? The answer is Illinois. This is not because Unique's articles of incorporation list its address as 350 Randy Road, Carol Stream, DuPage County, Illinois. That is merely its registered address, which every corporation incorporated in Illinois must have. It need not be its principal executive office; it could be a mail drop or an attorney's office. To deem it the principal executive office would be to rewrite section 6323(f)(2). S. D'Antoni, Inc. v. Great Atlantic & Pac. Tea Co., 496 F.2d 1378, 1382-83 (5th Cir.1974). The following facts, however, taken together, show that this address really is (or rather was) that of Unique's principal executive office. Unique's president, Olsen, and the two other officers of the company, who were his relatives, had offices at this address, along with three part-time clerical employees. Most corporate financial records, and the corporation's minute book, were kept there. This was the only address listed on the corporation's stationery. The factoring agreement between Unique and Dimmitt & Owens lists Carol Stream as Unique's address, as do Dimmitt & Owens' own internal records. A post office box in DuPage County was listed on Unique's federal tax returns. DuPage County was both the obvious place for Dimmitt & Owens to search for tax liens against Unique and for the Internal Revenue Service to file such liens.

Whether in some more profound, more ultimate sense it was Unique's principal executive office is unanswerable. Unique's principal asset was a plant in California that manufactured plastic binders. Although the office in Illinois was spacious (7,500 square feet), its total payroll expense was only a small fraction (less than 3 percent) of the company's total payroll, the rest being incurred in California. Olsen, the presiding genius of the corporation, spent most of each week in California, returning to Illinois (where he and his family maintained their home) on weekends. During the week he slept in an apartment at the plant. Hence, Dimmitt & Owens argues, most of the direction of the company was supplied in California. Indeed, it argues that Unique's principal executive office was in Olsen's head, and since the head spent more time in California than in Illinois, California is where the company's principal executive office was. But this ingenious reasoning overlooks the purpose of section 6323(f)(2), which is the only sure guide to the meaning of "principal executive office." Neither the government nor lending institutions can be charged with knowledge of the details of a businessman's behavior. The statute asks for the location of an office, not of an officer. To all outward appearance, Unique, like many other companies, had its plant in one state and its headquarters in another. And outward appearances are what count in a statute designed to fix the simple, the natural, the obvious place for filing and searching for liens. That was Illinois in this case. Dimmitt & Owens failed to search for liens in Illinois, and must pay the price. Resolution of the issue on summary judgment was proper. The facts are uncontested; only the characterization is in issue.

The next question relates to the default judgment against the government. It came about in this way. On July 7, 1983, the district judge made his ruling in the government's favor and scheduled a conference for October 11 at which the government was to present a judgment order for his signature. Because the attorney from the Tax Division of the Justice Department in Washington who was handling the case had not entered an appearance, the judge's ruling was sent only to the assistant U.S. attorney in Chicago who had filed an appearance. This lawyer, in an impressive demonstration of bureaucratic rigidity, did not forward the ruling to the Tax Division, because he does not forward copies of decisions to it unless expressly requested to do so, which he had not been in this case. On February 1, 1984, the Tax Division's attorney on the case called the clerk of the court to find out about its status, and only then discovered that on October 11 the district judge had granted Dimmitt & Owens' motion for a default judgment when no one from the government had shown up at the conference. The next day that attorney submitted the Rule 60(b) motion to set aside the judgment.

Rule 60(b)(1) empowers the district judge to set aside a judgment (including a default judgment, see Rule 55(c)) within one year of its entry, because of mistake or inadvertence. In almost every reported case involving a motion to set aside a judgment, the party against whom the judgment was entered was complaining of the judge's refusal to set it aside. The reason is that where the motion is granted, usually the order is not a final order but merely a preliminary to a trial of the case, and hence it is not reviewable till the trial is concluded. By that time, however, the question whether the default judgment should have been left alone is pretty academic; either the party who got that judgment has won, or, if he has lost, this shows that the default judgment really was unjust--a full trial has proved it so. See 10 Wright, Miller & Kane, supra, Sec. 2693, at pp. 476-77.

In the more familiar battleground of appeals from refusals to set aside default judgments, this court has moved away from the traditional position (see id., Sec. 2693) that such judgments are strongly disfavored; we are increasingly reluctant to reverse refusals to set them aside. See, e.g., Tolliver v. Northrop Corp., 786 F.2d 316 (7th Cir.1986); C.K.S. Engineers, Inc. v. White Mountain Gypsum Co., 726 F.2d 1202, 1205-06 (7th Cir.1984); United States v. DeFrantz, 708 F.2d 310 (7th Cir.1983); Breuer Electric Mfg. Co. v. Toronado Systems of America, Inc., 687 F.2d 182 (7th Cir.1982). This tendency reflects the overworked condition of the federal courts, including this court and the district courts of this circuit, a condition that naturally makes us reluctant to insist on the resurrection of deceased lawsuits. But the first and strongest line of defense against abusing the broad rights of access that modern claimants have to the federal district courts lies with those courts, so when a district judge decides to overlook a default and restore the case to his docket we are disinclined to interfere. See Duling v. Markun, 231 F.2d 833, 836 (7th Cir.1956). Moreover, the universally accepted rule that an order denying a motion to set aside a default judgment will be reversed only if the district judge can be said to have abused his discretion, see, e.g., Tolliver v. Northrop Corp., supra, at 318-319; Bermudez v. Reid, 720 F.2d 748 (2d Cir.1983), seems transferable intact to the case where the judge grants the motion. In either case the judge is weighing imponderables--the burden on his docket (which is to say the inconvenience to other litigants), the disturbance of expectations legitimately created by the default judgment, and the inroads on the general and essential principle that litigation must end, on the one hand, and on the other hand the injustice of allowing the default judgment to stand, which in turn is a function of both the merits of the movant's substantive claims and the strength of his excuse for committing the default. With the standard of decision so multifaceted, the appellate court's ability to fault the district judge's application of the standard is quite limited, and the scope of effective judicial review is therefore slight.

The judge in this case did not abuse his discretion, which is to say did not act unreasonably, in deciding to forgive the government's default. Since he had already ruled in the government's favor, it was clear that both a considerable injustice would be committed if the default judgment against the government were allowed to stand and that setting aside the judgment would not burden the district court with a trial or other extended proceedings, the merits of the lawsuit having already been resolved. In these circumstances and in the absence of any showing that Dimmitt & Owens was hurt because it relied on the default judgment during the three and a half months in which it was in effect before the motion to set it aside was filed, the district judge did not abuse his judgment in granting the motion, merely because the mistake that had led to the default judgment was a particularly stupid one. When we consider that innocent taxpayers (as we may, with some poetic license, describe the beneficiaries of successful efforts by the Internal Revenue Service to enforce its rights) will be out some $300,000 if the default judgment is enforced, that mistakes in communication even within the same department of the vast federal bureaucracy have been made inevitable by the modern growth of government, and that the government was made to compensate Dimmitt & Owens for the legal fees that the latter incurred in fighting to hold on to the default judgment, we cannot say that the district judge did the wrong thing in setting aside the judgment. Of course an argument can be made that taxpayers and everyone else will be better off if the government learns to avoid these mistakes and that enforcing the default judgment will help it to do so more than our criticisms or the district judge's forcing it to pay the modest attorney's fees (less than $4,000) that Dimmitt & Owens incurred in resisting the Rule 60(b) motion. But we do not conceive ourselves to have the authority to insist on so draconian a sanction. Rule 60(b) gives the district judge a power of lenity that he did not abuse in this case.

We need not decide whether the default judgment itself was invalid because of Fed.R.Civ.P. 55(e), which provides that no default judgment shall be entered against the United States "unless the claimant establishes his claim or right to relief by evidence satisfactory to the court," which Dimmitt & Owens manifestly failed to do, the court having in fact ruled in the government's favor on the merits of the suit. See 10 Wright, Miller & Kane, supra, Sec. 2702. The problem is that although in form a suit by Dimmitt & Owens against the United States, in substance the suit was, or at least became, a suit by the United States to foreclose its tax lien. That must be why, when the district judge entered a judgment of default, it was a judgment dismissing the case rather than awarding the injunction that Dimmitt & Owens had sought against the levy. The reference in Rule 55(e) to the claimant's "claim or right to relief" suggests that the rule may be limited to cases where the default judgment is in favor of a claimant against the government, and thus may not cover a case where the government is the plaintiff. However, in United States v. Geisler, 174 F.2d 992, 999 (7th Cir.1949), this court assumed that the rule does apply in such a case, and we have found a similar assumption expressed in dicta in two cases from other circuits. See United States v. Balanced Financial Management, Inc., 769 F.2d 1440, 1450 (10th Cir.1985); United States v. Sumitomo Marine & Fire Ins. Co., 617 F.2d 1365, 1370 (9th Cir.1980). But we need not determine in this case the scope of Rule 55(e). The judge did not purport to rely on Rule 55(e), and Rules 55(c) and 60(b)(1) gave him adequate authority to set aside the default judgment.

The other issues raised by the appeal merit only brief discussion. Dimmitt & Owens contends that it should not be liable for tax penalties, as well as unpaid taxes, assessed against Unique. But 26 U.S.C. Sec. 6662(a) provides that additions to tax, specifically including tax penalties of the type assessed against Unique, shall be collected in the same manner as taxes, while section 6321 provides that the amount of any tax lien shall include not only the unpaid tax but also assessable penalties. Finally, the evidence on which the amount of the judgment was computed was sufficient.

AFFIRMED.