Citi's Steven Englander On The USD Impact Of A Second Homeland Investment Act

The concept of a foreign tax repatriation holiday is nothing new to Zero Hedge, and has been discussed extensively before (here and here). There are those who believe that this "holiday", should it be allowed by Congress, will have a far greater stimulus on the economy than the ongoing "QE to infinity" rolling fiat destruction, which does nothing for real asset values and merely revalues prices in nominal terms. Today we present the thoughts of Citi's FX strategist Steven Englander on the topic of a second "Homeland Investment Act" which soon enough may be the last trump card to jump start a once again sputtering "recovery" despite the ongoing impact of QE2 and the unwind of the SFP program. And just like Goldman 10 days ago, Citi is the second major bank to reach the conclusion that louder whispers of a HIA2 will be broadly dollar positive (something to keep in mind considering net USD spec non-commercial bets are at multi-year bearish positions), with the following key differences from HIA1: "1) Be more front-loaded; 2 ) Result in much higher volumes; 3)… And maybe a higher USD component; 4) But will result in more selling flows from reserve managers into the buying by the private sector."

HIA-2 under discussion

  • HIA is attractive as a way of reducing effective corporate taxes temporarily and improving the tone of the US corporate sector, but the direct impact on investment and employment appears limited
  • The total flows are likely to be much higher than in 2005
  • The non-USD share is less certain but may be somewhat lower
  • Central banks may see this as a golden opportunity to diversify

A renewed program to allow repatriation of foreign profits at favourable tax rates is again under discussion in the context of broader corporate tax reform. Proponents argue that it provides inexpensive stimulus to the US economy at a limited budget cost. Opponents argue that it provides few practical benefits; rather it creates incentives to keep earnings abroad in anticipation of subsequent rounds of HIA (Homeland Investment Act – the actual name of the bill was the American Jobs Creation Act of 2004, but we will use HIA-1 to refer to the 2004 bill and HIA-2 to refer to any prospective 2011 measure).

Below we argue that the HIA-1 was oversold in terms of its direct macroeconomic stimulus, but probably had a bigger punch indirectly through improved asset market conditions. The HIA-1 impact on the USD is still debated, but in retrospect the direction was unambiguously USD positive, both via the direct repatriation of earnings which had been held abroad in foreign currencies and probably through a broader positive impact on the attractiveness of US companies and markets.

Currently the sums held abroad appear to be much larger than was the case during HIA-1, possibly as much as twice as large. HIA-2 would generate major tax benefits for the companies with un-repatriated overseas earnings and likely generate significant repatriation of earnings. The USD impact is again likely to be very positive, with the caveat that a doubling of the size of repatriated earnings versus 2005 does not necessarily mean that there will be twice as much USD buying.

The extent to which there is a USD effect will depend on whether the un-repatriated earnings are kept in foreign currencies or in USD. In practice, this often depends on whether the functional currency of the US subsidiary is USD or local currency and whether the earnings were kept abroad in anticipation of ultimate reinvestment locally or in anticipation of a subsequent HIA program. Nevertheless, again it seems very likely that there would be a significant USD positive effect from HIA-2, although unlike in 2005, we could see significant selling by reserve managers into the USD buying.

The debate

The pros and cons of HIA are well known and much discussed. Nevertheless, it is useful to summarize the arguments on both sides if only to be able to gauge whether the debate favours passage or not. The USD impact has not been part of the debate, with investors and policymakers correctly seeing the USD outcome as reflecting broader economic and asset market effects rather than being a focus on its own.

The disadvantages:

  1. In 2005, HIA-1 delivered much less in direct employment and investment than promised. For example, see “Tax Incentives and Domestic Investment: An Empirical Analysis of the Repatriation Decisions of U.S. Multinational Corporations Following the Implementation of the Homeland Investment Act of 2004” Michaele L. Morrow, Ph.D,. Dissertation, Texas tech University, May 2008, or “Watch What I Do, Not What I Say: The Unintended Consequences of the Homeland Investment Act “ J Dhammika Dharmapala, C. Fritz Foley and Kristin J. Forbes, Journal of Finance, forthcoming (2011).
  2. Under HIA-1, the incentives to increase employment and investment were limited. The major impact of HIA-1 was to allow foreign earnings to be repatriated at low tax rates, with few binding additional requirements. From firm’s point of view, HIA-1 was equivalent to a lump-sum tax benefit which would generate additional investment and employment primarily in cases in which firms had restricted access to credit markets. Firms with large amounts of profits abroad probably could borrow domestically for hiring or capital expansion so would not have been constrained in their prior investment decisions.
  3. Crafting a bill that increases direct marginal incentives for employment and investment is difficult. If the requirements are too stringent, firms will simply pass on repatriation. If firms are already unconstrained with respect to hiring and investment, a marginal increase may bring forward investment plans into 2011, with some payback in subsequent years. If the terms are relatively lax, as in HIA-1, the impact on direct employment and investment will be small.
  4. The firms that have the money abroad (tech, pharma) are not the sectors that need the most balance sheet help (households, real estate, state and local government) nor does it help firms whose operations are primarily domestic.
  5. Repeating HIA produces incentives for firms to keep funds abroad. There is the risk that firms will see HIA as a once or twice a decade low-tax repatriation opportunity. The extent of these incentives depends on the gap between US domestic and foreign tax rates. The combined effect of HIA plus a reduction in US corporate rates would largely mitigate these incentives. Surprisingly, BEA data suggests that until the possibility for HIA-2 emerged again in early 2009, the aggregate dividend repatriation rate was not much lower than it had been prior to HIA-1(Figure 1), and the low repatriation since 2009 could also reflect limited US investment possibilities.

The advantages:

  1. HIA-2 presents an opportunity for the Obama Administration to demonstrate its commitment toward a more business friendly approach to an important constituency.
  2. HIA-2 eases access to funds that are viewed as locked abroad to some degree. A corporate tax system that encourages firms to keep cash abroad while borrowing domestically is arguably less than optimal.
  3. The sums involved are substantial. There are estimates of up to USD 1 trn kept abroad - roughly half the cash currently held by US corporates. Data from the BEA shows USD1.2 trn of un-repatriated earnings since 2006, significantly more than had been accumulated over the 1990-2004 period (Figure 2).
  4. The tax costing can be relatively benign because the low tax rate is largely offset by the increase in flows. The repatriation flows in response to the lower corporate tax rate are so high that they largely pay for themselves (in subsequent years, costing depends on how much flows are expected to be reduced by anticipation of future HIA)
  5. In contrast to 2005, improving balance sheets and financial statements is higher on the list of policy priorities. One of the Fed’s stated objectives in QE2 was improving the attractiveness of other asset markets relative to the bond market, so in 2011, balance sheet improvement can be viewed as a macroeconomic policy goal.
  6. 2005 was one of the best years of the decade in terms of asset markets and growth so indirect effects may have been large. It is hard to pin down these indirect effects (and obviously there were broader macroeconomic forces at play) but 2005 was a year of strong employment and investment growth (Figure 3), a strong USD, and sharply revised expectations of how quickly the Fed could normalize rates. From the time the bill was passed in late 2004 till the end of 2005, expectations of Dec 2005 short rates rose from just over 3% to 4.5% (Figure 4).
  7. No one’s ox is gored, at least not directly. It is difficult to craft a stimulus package that is relatively cheap in budget terms and which provides broad stimulus and that does not carry a well-defined set of losers. Especially if combined with broader corporate tax reform that narrows the gap between US and foreign corporate tax rates, HIA-2 may be viewed as more attractive and practical than other more theoretically attractive stimulus measures.

Impact on the USD

In 2005, the debate was focused on the amounts that would be repatriated and the extent to which the earnings held abroad were already in USD. There was also considerable scepticism with respect to the size of the impact that repatriation flows would have on the USD. There is still some debate focused on the degree to which funds these funds were held in USD.
We are of the view that HIA-1 was a USD positive and that it did generate USD-positive flows. That was clear anecdotally and it seems to be confirmed by price action although confirmation in official data is difficult to find.

In the event that HIA-2 becomes a reality, we expect that it will be USD positive but the impact will differ from HIA-1 in that the impact will:

  1. Be more front-loaded.
  2. Result in much higher volumes of repatriation…
  3. … And maybe a higher USD component
  4. But will result in more selling flows from reserve managers into the buying by the private sector

What we saw in 2005 was that the timing of repatriation flows could be erratic. Some firms that were very involved in the passage of the bill moved early in 2005. However, many firms moved slowly on HIA because the Treasury did not give full guidance on interpreting the provisions until early May. Some corporate boards were reluctant to sign off on the plan (given their liability under Oxley-Sarbanes) until they were sure that they were in full compliance. In the BEA data on earnings repatriation, a somewhat larger than normal share of repatriations occurred in the second half of 2005 - so the uncertainty may have had an effect on flows.) If HIA-2 becomes a reality, the odds are that firms would be quicker off the mark to repatriate and that investors would be quick off the mark to price in HIA effects.

We have completed an informal analysis of the financial statements of about a dozen firms that repatriated earnings in 2005. We compared the amounts that they repatriated in 2005 with the amounts that they currently indicate are permanently invested abroad in their most recent financial statements. (The firms are not randomly chosen in the statistical sense. However, we had no prior knowledge of how their current un-repatriated earnings compared with 2004).

We find that the numbers are now 80% higher. Although this is a small sample, we have no reason to think the numbers are not indicative at least that the sums involved are large. There are two dimensions to this increment. The obvious one is that HIA-1 was very attractive from the corporate point of view and that firms may have been conscious that another HIA might ultimately have been in the pipeline.

Less obvious is that many firms had only modest benefits from HIA-1. The HIA-1 bill specified that firms either had to have declared the amount of earnings that were permanently reinvested abroad or they had to provision for US tax liabilities. Otherwise, they could only bring back USD 500mn. Many firms that found themselves able to repatriate only USD500mn in 2005 have significantly ramped up their declaration of earnings permanently reinvested abroad.

The second issue is the USD share. This share may be less than in 2005 (although that is not guaranteed). If US subsidiaries abroad use USD as their functional currency or if they were keeping earnings abroad in anticipation of HIA-2, the odds are that a significant share is in USD. If this is the case, there would not be direct FX implications (although attractive US asset markets are themselves an inducement for capital inflows). If the firms were viewing local investments as the more likely investment direction, the odds are higher that the funds would be kept in local currency.

This all sounds USD positive. The one caveat is that reserve managers might see HIA-2 as an opportunity to sell into a sizeable round of USD buying by US corporates. From the perspective of a long-term investor who is overweight USD, it may be tempting to take the other side of what is clearly a temporary wave of one-off flows. The IMF COFER data show only limited USD accumulation between 2003 and 2007, and some evidence that diversification into other currencies was more aggressive than was possible in years when the USD was under continuous pressure. While such USD selling by reserve managers is unlikely to fully reverse any HIA-2 induced buying, there is the risk that USD supply and demand may be more evenly matched than investors expect or was the case in 2005.