100 years ago, 1905 was the "annus mirabilis", the miraculous year when Albert Einstein made some of his major and most revolutionary scientific discoveries, some of which - especially the special theory of relativity -  later led to the general theory of relativity; thus, a new view of the laws of gravity emerged that radically challenged the traditional Newtonian laws of gravity. 100 years later, in 2005, it appears that the US dollar is defying the economic laws of gravity and putting them into doubt: while the US current account deficit has grown larger and larger - from $665b in 2004 to a predicted $800b plus this year - the US dollar has been strengthening throughout the year. Does this mean that we should rewrite the laws of economic gravity, i.e. the idea that - over time - a large current account deficit will be associated with a falling currency value? The simple answer is not and I will try to explain why.

First of all, the relation between the current account and the exchange rate is complex. If we do believe that large current account deficits are - at times - caused by an appreciating currency, then in such periods the current account will be worsening while the currency is appreciating. In these periods, the causality goes from the exchange rate to the current account; for the US dollar such were the periods going from 1995 to 2002 and, most recently, 2005. Of course, at some point the law of currency gravity starts to reemerge and a large current account deficit causes a depreciation of the currency to induce  - over time - an improvement of the current account. In these periods, as in 2002-2004, a large current account deficit causes a currency depreciation. So, current account deficits may be associated with an appreciating or depreciating currency depending on the causality chain.

The issue is thus: since a current account deficit should - like the law of gravity - lead to a falling currency, what are the forces that - at least in the short run - allow a currency to defy such laws of gravity and levitate in spite of a large current account imbalance?

The balance of payment is the sum of the current and capital account where the sum of those two components is equal to the change in the foreign currency reserves of the central bank. Since the US rarely intervenes in the forex market, the exchange rate is determined by the current account and the capital account. The relation between the current account and the currency value is simple: as long as the capital account does not lead to an inflow of capital to finance a current account deficit, such a deficit should lead to a weakening currency. In fact, the demand for foreign currency coming from imports of goods and services and the payments of factor incomes from foreign ownership of domestic assets dominates the supply of foreign currency coming from export of goods and services and the income earned on the payments of factor incomes from the domestic ownership of foreign assets. This is the simple law of currency gravity.  Of course this is only half of the medallion as, for any given current account deficit, the capital account determines, at least in the short run, whether there is enough foreign inflows of capital entering the domestic economy and increasing the demand for domestic currency and domestic assets. In the capital account such inflows tend to increase the demand for the domestic currency and lead to a currency appreciation; while the capital outflows lead to a reduced demand for domestic currency and domestic assets. They thus lead to an increased demand for foreign currency and lead to a currency depreciation.

Thus, for any given current account deficit, if the supply of financing from the capital account is large and strong enough, the currency will tend to appreciate - driven by net capital inflows - and one would thus observe a current account worsening. Conversely, if the supply of financing from the capital account is low and falling, the currency will tend to depreciate - driven by net capital outflows - and one would thus observe the current account improving over time as the currency depreciates. Which factors determine the capital account and net capital flows? Relative short term and long term interest rate differentials; political risk factors that affect risk premia; relative returns on equity markets that affect the relative attractiveness of domestic versus foreign equities; the relative GDP growth rate differentials that affect both expected future short rates but also relative returns on equities and the risk premia.

So, in 2005, if the dollar had temporarily defied the laws of gravity coming from the large, heavy and growing gravitational force of a worsening current account deficit, it has done so as the capital account and the factors affecting the capital account were supportive of the dollar.

So, what are the capital account factors that have contributed this year to the anti-gravitational forces that caused a rising dollar and, thus, a worsening current account deficit?

First, the increasing differential between short term interest rates in the US relative to Europe and Japan as the Fed has kept on tightening while the ECB and the BOJ have remained on hold. Note however that, according to the UIPC (Uncovered Interest Parity Condition) a US tigthening strengthens the US dollar on impact but it does lead, over time to an expected and actual depreciation of the dollar as the interest rate differential is a return to compensate for the expected fall of the US. So, unless the US keeps on tightening, a stable interest rate differential would be dollar bearish.

Second, the increase in the relative growth differential between the US and Japan/EU as the US has kept on growing around potential while Eurozone and Japanese growth was sub par.

Third, the effects of the Homeland Investment Act (HIA) that has led to the return to the US of almost $200b of US profits that were kept abroad for tax reasons.

I will argue that the issue in 2006 is not whether the dollar will be falling but rather when and how much as the gravitational forces that would weaken the dollar - a large current account deficit - will weigh with even greater force while the antigravitational forces that have lifted the dollar in 2005 will fizzle out over time. Specifically:

- The short term interest rate differential will reverse against the US as the Fed will eventually stop its tightening cycle (and possibly reverse it if US growth sharply slows down) while the ECB will start its tightening phase and the BOJ will drop its ZIRP (Zero Interest Rate Policy rather than its "zip").

- The growth differential factor will also turn against the US if the US growth rate slows down towards 2%-2.5% (driven down by a shopped out consumer with negative savings, high oil prices, the confidence effects of Katrita, high debt and debt servicing ratios and a fizzling housing bubble) while Japanese growth may recover towards potential (2%) and even the anemic Eurozone growth rate may recover towards a 2% potential level.

- The dollar boosting effects of the HIA will disappear as this profit-capital returning factor will be phased out by the expiration of this tax incentive.

- The relative returns in equity markets - that already in 2005 saw the outperformance of European and Japanese equities relative to US equities - may get reinforced if US growth slows down while Japanese and Eurozone growth recovers.

- The relative yields on long term bonds may also become bearish for the dollar as capital losses on US dollar long term fixed income assets coming from rising US long rates may be - for a while - larger than similar losses on Eurozone and Japanese bonds.

- The attractiveness of US dollar assets may be reduced by a bursting - or even flattening out - of the US housing bubble compared to Japan and the Eurozone where such bubble did not materialize (apart from specific exceptions such as Spain). The capital gains from such a bubble have sustained, so far, the demand for US dollar assets.

- The political factors that weakened the Euro in 2005 - the EU constitutional referenda failures, the EU fight over the budget, the mixed results of the German election,  the French riots - may not be as serious as in 2005 as some of this issues will be tackled by Europe and resolved. 2006 may be the year when a range of pro-reform governments will be elected and/or strengthened in Italy, France and Germany. Conversely, the US political factors may weaken the dollar as the US administration looks like lame duck even this early in its second term; Iraq is becoming an increasing quagmire; and other serious domestic (Katrita and its fiscal effects) and international challenges (geostrategic stress points related to terrorism, Iran, Iraq, North Korea, etc.) may fester and worsen.

- China may decide to move its currency by about 10% or more in the next 12 months, thus leading not only to a weakening of the dollar relative to the RMB but also to an appreciation of a wide range of Asian currencies - including the Yen - relative to the US dollar. China would move for two reasons: threats of US protectionism as the Chinese trade surplus has surged; the domestic need to cool down the overheated economy.  The reduction in the relative demand for US dollar assets deriving from a lower forex intervention by central banks and stronger Asian currencies will then affect the dollar directly but also indirectly as the private carry trade Asian demand for US dollar asset is not independent from the currency risk that appreciating Asian currencies will imply.

Thus, all in all, all the major factors listed above would tend to weaken the dollar. The threats to this bearish dollar scenario may come from the following sources: first, the Fed may not stop tightening for some time, while BOJ may stay on hold with ZIRP for longer than expected while ECB may tighten only slowly. Second, Japanese and Eurozone growth may disappoint while US growth may not decelerate as much or, if US growth decelerate a lot, other G7 growth rates would also decelerate rapidly with collateral effects on relative short and long term interest rates and returns on equities. Third, inter-Eurozone political problems and division could fester as the EU integration process may slow down.  Fourth, China may resist a meaningful currency move. But, all in all, I believe that in 2006 the relative evolution of these factors will tend to move against the US dollar rather than strengthen it as I would give little probability to the events described above that may slow down the weakening of the US dollar.

In conclusion, the behavior of the US dollar in 2005 does not require a rewriting of the laws of Newtonian or Einstenian gravity as in 2005 the cyclical antigravitational forces supporting the dollar dominated the gravitational forces of a worsening current account that would have - coeteris paribus - weakened the dollar. In 2006 the structural medium term factor that will tend to weaken the dollar - the large and growing US current account deficit - will reassert its role while the short term cyclical factors that have lifted the dollar this year will tend to weaken their effect. So, at the end you cannot fight the laws of gravity as the cyclical forces that have defied such gravity are temporary while the forces that will cause a gravitational fall of the US dollar are as strong as ever.