Yesterday at the NYU Tax Policy Colloquium, Victor Fleischer presented an early draft of his paper, "Tax and the Boundaries of the Firm." The paper discusses an interesting set of issues pertaining to how the features of U.S. corporate taxation affect firm "size," such as with respect to the Coasean tradeoff between make-or-buy decisions.
Rather than writing a fresh comment on the issues raised by the paper, why don't I offer here an expanded version of the outline that I prepared to help guide discussion at the session:
1. Effects of current tax law, absent regulatory arbitrage
(a) What is firm size, and when is it a Coase story?
(i) Multi-dimensional (e.g., make-or-buy, assets inside or outside the firm, multi-industry conglomerates, full-time vs. part-time vs. independent contractors)
(ii) What relationship to issues of debt vs. equity choice?
(b) Significance of transfer pricing
(i) Only matters if the “pocket” does (firms, households) – e.g., international transfer pricing, domestic manufacturing deduction
(ii) Irrelevant to well-designed VAT or retail sales tax; can matter in domestic income tax due to the “cascading” tax on inter-firm transactions (from inclusion on one side, capitalization on the other) if not replicated by in-house capitalization rules
(iii) International: pure WW versus territorial (see example at the end of this document)
(c) Separate businesses (loss nonrefundability creates incentives for conglomeration)
(d) Agglomeration (i.e., excess firm size because earnings are retained to avoid the tax on shareholder distributions)
(i) Wouldn’t it be anticipated on the way in?
(ii) New view: uniform distributions tax does not induce lock-in
(iii) International new view: same as domestic, but permanently avoiding the tax may be more credible because the taxable “distributions” are intra-firm
(e) Full-time employees (vs. part-time, independent contractors) – note that companies often want to avoid having full-time employees despite the opportunity to give them tax-free, in lieu of taxable, compensation
(f) Choice of tax rate – any differences between the corporate and individual rates may affect firm or entity choice; the ease of taxable income-shifting through partnerships can be a reason for using them
(g) Debt versus equity
(i) Modigliani-Miller: choice is irrelevant apart from bankruptcy, tax, asymmetric information / agency cost issues
(ii) Tax shield vs. bankruptcy risk tradeoff
(iii) Tradeoffs from agency costs / asymmetric information: debt better in some scenarios, equity in others.
(iv) Miller equilibrium: debt vs. equity is simply an election to pay tax at the corporate rate (via equity) or at one’s own rate (via debt)
2. Regulatory arbitrage
(a) General issue is costly vs. cheap electivity
(b) Examples: check-the-box, anti-avoidance rules, debt vs. equity, rules constraining realization and recognition
(c) Transaction cost tradeoff: more/cheaper vs. fewer/costlier transactions
(d) Need to assess merits of the particular taxes that regulatory arbitrage permits one to avoid (e.g., for tax shelters, debt-equity under Miller, or domestic vs. international check-the-box)
3. Policy implications
Corporate integration? Lowering the corporate tax rate? International?
Make versus buy and the U.S. international tax rules
The U.S. tax rate is 35%, that in Ireland is 12.5%. A U.S. company will use parts made in Ireland (a decision that itself may reflect this tax rate difference) but is choosing between make and buy (i.e., subsidiary vs. arm’s length purchase).
Suppose that separately owned U.S. and Irish firms would earn $16 and $8, respectively. After paying national taxes, they collectively have $10.40 + $7 = $17.40.
(a) Territoriality as inefficiently encouraging “make”
Suppose merger reduced their combined income to $22, but that they could report this as $6 in the U.S. and $16 in Ireland. Now the after-tax is $3.90 + $14 = $17.90.
(b) Pure worldwide as inefficiently encouraging “buy”
Suppose merger increased their combined income to $26, all subject (after foreign tax credits) to U.S. tax. Transfer pricing is irrelevant, but now the after-tax is just $16.90.
(c) Deferral – In principle an indeterminate tradeoff between (a) and (b); in practice we can be confident that (a) is more significant.