Untaxed at Marriage, Taxed at Divorce: Who Pays When a Cross-Border Joint Venture Ends

By Douglas S. Stransky, J.D., LL.M.  | Partner, Sullivan & Worcester LLP

Joint ventures get formed in a hopeful mood and hopeful moods make for thin tax provisions.

Consider a hypothetical drawn from recurring patterns in my practice, not from the facts of any particular client matter.

Corliss Industrial, a U.S. manufacturer, forms a 50/50 joint venture LLC with a German strategic partner. Corliss contributes a plant and the process technology that runs it, worth $40 million against $8 million of tax basis. The German side contributes $40 million of cash to fund expansion. It does not, however, hold its interest directly. Like most non-U.S. investors in a U.S. LLC, it forms a U.S. corporate blocker to be the partner because its home country may not view the LLC as transparent and the credit mechanics for tax paid on an allocable share of partnership income do not line up across the border. The blocker is a routine piece of inbound structuring, and once it is in place nobody gives it further thought.

Internal Revenue Code Section 721 leaves Corliss’s contribution untaxed, the deal teams confirm that much, and the LLC agreement disposes of the rest in a single sentence: allocations with respect to contributed property shall be made under Section 704(c) using any method permitted by the regulations, as determined by the managing member.

The champagne is opened. Two tax clocks start ticking, one economic bargain gets struck silently and the blocker has consequences no one has modeled. None of this is hidden. Every one of these issues is visible at formation to advisors who go looking. The trouble is that the deal’s momentum runs the other way: the contribution is untaxed, the term sheet is agreed, and the tax article gets assembled from the last deal’s precedent.

Why the boilerplate is a bargain in disguise

Start with the silent bargain. Corliss brought $32 million of built-in gain into the venture, and Section 704(c) exists to make sure that gain is eventually taxed to Corliss rather than shared with the partner that paid full value. What the boilerplate concealed is that the regulations offer several methods for accomplishing this and for depreciable property the methods produce meaningfully different outcomes.

Under the traditional method, a limitation known as the ceiling rule can leave the non-contributing partner with smaller depreciation deductions than its economic investment supports. The remedial method cures that distortion with offsetting notional allocations, which is better for the cash partner and worse for Corliss. Here the blocker matters. The German side’s partner is not a foreign investor with limited U.S. tax appetite; it is a U.S. corporation paying full U.S. corporate tax on its distributive share and every depreciation dollar the ceiling rule takes away is a real cost inside that blocker. The method choice moves money between the partners for years. In a well-advised deal it is negotiated at the term sheet stage along with the other economics. In many deals, though, the sentence arrives from precedent and stays: the method is left to the managing member and whoever controls the tax function later discovers a lever the other side never knew it was holding.

The seven-year clocks

Now the clocks. The partnership rules contain what practitioners call the mixing bowl provisions, and they exist because Congress noticed that a partnership could otherwise be used to swap appreciated assets tax free. Two sections do the work. Under Section 704(c)(1)(B), if the property Corliss contributed is distributed to the other partner within seven years of contribution, Corliss recognizes its remaining built-in gain at that moment. Under Section 737, the mirror rule, a distribution of other property to Corliss within the same window can likewise trigger its precontribution gain.

Seven years sounds like a long time. It is not. Joint ventures restructure constantly and the restructurings arrive in year four or five with commercial logic behind them: the partners want to separate the business lines, or move the technology closer to the European operations, or redeem part of one side’s interest to rebalance ownership. Each of those transactions can involve a distribution and each tests whether the formation documents answered a question the parties did not know they were being asked.

In the hypothetical, year six brings the divorce. It is amicable: the German side takes the technology and Corliss takes back the plant plus cash. Under Section 704(c)(1)(B), though, the distribution of the technology is, for Corliss, a taxable event to the extent of the remaining built-in gain in that asset.  That gain was not forgiven at the marriage, only deferred, and the divorce is when it comes due. And the German side now confronts its own structure. A distribution of the technology reaches only the blocker, because the blocker is the partner. Moving the asset out of a U.S. C corporation and into the German group is a second taxable event at the corporate level, with withholding on any dividend that follows. The structure that solved a classification problem at formation has added a level of tax at the exit. Both partners are now paying for decisions made in the optimistic months.

The exit nobody modeled

Suppose instead the venture succeeds and the parties simply part by sale. Corliss’s side is familiar territory: gain on the interest, with a portion recharacterized as ordinary income under Section 751 to the extent of the venture’s hot assets, which in a manufacturing business with inventory and depreciation recapture is rarely trivial.

The German side’s exit runs through the blocker, and the usual sequence is well worn: the blocker sells its LLC interest, pays U.S. corporate tax on the gain including its own Section 751 component and then liquidates, distributing the after-tax proceeds to its German parent. Done in that order, with the blocker holding only cash and the liquidation qualifying under Section 332, the liquidation itself generally draws no further U.S. tax at the shareholder level, which is precisely why the blocker was tolerable in the first place; one level of U.S. corporate tax was always the price of admission, and the structure is built so that it is also the whole price. The caveat that earns its place in every one of these deals is FIRPTA. A venture that owns a U.S. plant puts real property on the balance sheet, and if the blocker is a U.S. real property holding corporation, the analysis of the parent’s stock and the sequencing of sale and liquidation both require care rather than reflex.

And where a non-U.S. investor does hold a partnership interest directly, without a blocker, a different set of rules applies at the exit, since Section 864(c)(8) can treat the gain as effectively connected and Section 1446(f) makes the buyer withhold 10% of the amount realized unless the right certificates are in hand. The structure changes which rules show up at the closing. It never changes the fact that some of them do.

What the agreement should have said

The remedy is unglamorous drafting at formation, when leverage is balanced and nobody is angry yet. The Section 704(c) method belongs in the agreement itself, chosen deliberately and priced into the economics. Distributions that would trigger the mixing bowl rules during the seven-year window deserve either a consent right for the contributing partner or an indemnity that puts the cost on the party that wanted the transaction. The exit provisions should presuppose the structures at the table, blocker included, with cooperation covenants and certificate mechanics written down while everyone is still friends. And the tax article should be read against the term sheet by someone asking a single question about every clause: when the partners eventually separate, restructure or exit, who pays under this sentence?

That question, asked early, is most of the game. Asking it well requires seeing the whole arc of a venture at once, from the contribution through the restructuring through the unwind, which is a perspective that comes from deals rather than from any single Code section.

Giving practitioners that arc is what I tried to do in International M&A and Joint Ventures: Key U.S. Taxation Issues, which devotes a full chapter to the U.S. tax considerations of joint ventures and pairs it with case studies that follow ventures like the Corliss hypothetical from formation to unwind, along with practice aids including sample tax provisions for transaction agreements. But formation is only the beginning. The chapters I most wanted to write are about everything that happens after the champagne.

International M&A and Joint Ventures: Key U.S. Taxation Issues

International M&A and Joint Ventures: Key U.S. Taxation Issues examines the U.S. tax implications of cross-border mergers, acquisitions and joint ventures from planning and due diligence through structuring, negotiation and post-closing considerations. The treatise includes practical analysis, case studies, tax due diligence checklists, sample acquisition agreement provisions and guidance on tax insurance and risk allocation to help practitioners navigate international transactions. The title is available in multiple formats on the LexisNexis® Bookstore.

About the Author

Douglas S. Stransky, J.D., LL.M. (Taxation), is a partner and the leader of the Tax Practice Group at Sullivan & Worcester LLP in Boston, where he has practiced for nearly two decades. He concentrates his practice on international tax planning, with a particular emphasis on cross-border mergers and acquisitions, joint ventures, and corporate restructurings. He is a Lecturer in Law in the Graduate Tax Program at Boston University School of Law, where he has taught corporate and international tax courses for over 15 years. Stransky is the author of International M&A and Joint Ventures: Key U.S. Taxation Issues, a LexisNexis treatise that examines the U.S. tax considerations involved in cross-border mergers, acquisitions and joint ventures.