Economics is relatively good at telling us what should happen next, but it's downright awful at telling us when next is. For example, if consumers keep borrowing more and more to spend more and more, at some point, economics says, the debt pyramid will come crashing down. But is "some point" the second half of 2005, 2006 or 2010?

If you study the nature of the lags that derailed the predictions of January 2005, that "some point" looks likely to be 2006. Maybe mid-2006.

Consider the mess that has swallowed the euro in the last few months. Slowing economic growth in Europe -- slower than in the U.S., for sure -- and lower interest rates had taken a 5% bite out of the euro by the end of March. And then came the dust-up as French and Dutch voters said "no" to a new European Union constitution. Knock another 5% off the euro.

And what has been bad for the euro has been good for the dollar.

It's not so much that the dollar looks so much better than it did in January. The U.S. still faces huge budget and trade deficits. In fact, the April trade deficit climbed to $57 billion after the monthly deficit had dropped to a revised $54 billion in March. But some of the fear that Asian banks would start dumping their dollars to buy euros has dissipated, now that so much uncertainty swirls around the euro.

We haven't yet seen a peak in worry about the euro, either. With a good chance that next week's European Union summit will end in bitter fighting between British Prime Minister Tony Blair and French President Jacques Chirac, and that the government of Germany's Gerhard Schroeder faces defeat in elections scheduled for the fall, it's hard to see the political turmoil that's now driving the euro's decline ending before the fourth quarter.

BNP Paribas now sees the euro falling to $1.16 by the third quarter from the current $1.22, before it begins a recovery to $1.28 by the first quarter of 2006. That seems a reasonable scenario, especially if the Federal Reserve keeps raising short-term interest rates in June and August to 3.5%, and the European Central Bank finally cuts interest rates to revive economic growth. Higher U.S. interest rates compared with those in Europe would support a stronger dollar.

Of course, a strong dollar, which hurts U.S. exports, and higher interest rates at home, which dampen domestic spending, do add up to lower economic growth in the U.S. Not all at once. And not by a huge amount. But the drop from 4% growth in the third quarter of 2004 to 3.8% in the fourth quarter to 3.5% in the first quarter of 2005 does put the economy on a path that leads to growth nearer 3% by 2006.

The Role of Globalization

If you go back to the predictions made at the beginning of 2005, the consensus was that short-term interest would hit 3.5% or so by the end of the year. That looks about right, with Alan Greenspan signaling last week that the Federal Reserve will raise rates at the end of June and in early August.

But predictions that long-term rates would hit 5.5% or even 6% by the end of 2005 stand to be dead wrong.

The globalization of the financial markets has changed the way U.S. interest rates respond to rate moves by the Federal Reserve. Look at one popular trade now: borrowing in Asia, where interest rates are even lower than they are in the U.S., and then investing the proceeds in the U.S. bond market. It's pretty profitable when the overnight U.S. dollar LIBOR rate is at 3.03% and the Japanese LIBOR stands at 0.03%, as it did on June 9. (LIBOR, the London interbank offered rate, is available only to the most creditworthy of international banks. But it's often the base rate used in calculating the yield on other loans, so it gives a good idea of the spread.)

Asia is awash with cash now, thanks to high savings rates and huge cash inflows from international trade surpluses. (That's the other side of the U.S. trade deficit.) This keeps interest rates low in these markets, and the huge movements of cash across international borders works, for now, to push U.S. interest rates lower.

Add in global demand for long-dated bonds, as private and public pension funds try to match their investments to their obligations to rapidly aging populations, and you start to get a possible explanation for the puzzle of rising short rates and falling long rates.

Euro Making the Dollar Look Good

This has the effect of driving down today's rates as well. But Greenspan recently told Congress that, this time, lower long-term yields may not be a signal the bond market is anticipating an economic slowdown. May not be. In other words, he doesn't know either.

The argument at the beginning of 2005 for a weaker dollar and higher U.S. interest rates was based, in good part, on a belief that international investors were becoming gradually unwilling to hold an ever-increasing supply of U.S. dollars. That unwillingness started to show up earlier this year in announcements from Asia central banks that they were thinking about reducing their portfolio concentration of dollars in favor of other currencies.

Haven't heard much talk like that since the euro's political crisis, have you? My best estimate now is that we won't see a significant upward trend in the yield on the U.S. 10-year Treasury until 1) the euro stabilizes and then starts to climb again vs. the dollar, and 2) the spread between the yield on the 10-year and two-year Treasury note vanishes. Today, the 10-year yields 3.9%, and the two-year yields 3.6%; this is an extraordinarily small spread. It's likely explained by the global need by pension funds for long-dated paper.

The Key Question

But I have to question how willing investors will be to take on eight more years of risk if the spread closes to 0.2 percentage points or vanishes completely. When short-term rates exceed long-term rates, it's called an inverted yield curve. The bond market traditionally gets very nervous when the yield curve inverts, because this has often signaled bad times ahead. There are still enough traditionalists in today's bond market to put the brakes on falling long-term yields when we get close to an inverted yield curve.

Where does that leave investors?

The predictions from the beginning of 2005 that look so wrong today are likely to still look wrong for the next six months: The dollar is likely to be strong for a while longer, interest rates don't look headed up at the long end, the bond market is more likely to move up than down, and economic growth will remain solid.

But as we move into 2006, those predictions from early 2005 are likely to seem more and more correct. The trends set by the economy, interest rates and currencies so far this year are closer to their end than to their beginning. And many of the strategies that have made money in the first half of 2005 will look increasingly risky as the year's second half plays out.

In other words, we market prognosticators weren't wrong in January 2005. We were just very, very early.