Nouriel Roubini Calls The Recession's End in Late 2009?

Nouriel Roubini has been one of the few noteworthy economists to be right about the housing market. And to be bearish well enough in advance to alert people before this crisis unfolded. I don't really follow Roubini except for browsing anything that might be reported on a few of my internet feeds but as I understand it, the majority of his analysis is based on a very U.S.-centric view of the housing crisis and its implications. I'm going to repeat a fact that I repeated often over the years for any new readers. There is no such thing as a consumer-driven recession.

Roubini has stated that he was misquoted recently about being bullish, which I obviously understand. But, he is clearly quoted in this Bloomberg article as stating the recession will end in twenty four months and that he anticipates the recession will end yet in 2009. We may technically see the economy eek out some stimulus-driven GDP in 2009 but this crisis is not going to end in 2009. As long as the Federal government is able to fund a war and a massive budget stimulus, there is obviously some point at which we will see year over year positive GDP. That is, assuming the private sector's GDP doesn't take another leg down such that government provided stimulus must reach even more astronomical levels. (Of course, it is surely a reasonable probability that it will.) That doesn't mean the economy is going to return to the economic levels of 2007. The economy will still be operating at substantially below trend until the next crisis hits. And there will be future crises. With a little bit of effort, a very short term forecast of positive GDP growth shouldn't be too difficult as long as we see some continued stabilization. Beyond the next month or two, accurate GDP forecasts are now nearly impossible because the global variables are too many and they are changing way too fast.

All of this thinking about an economic recovery, be it Roubini or otherwise presumes one point in particular. That we can curve fit the future onto the past. So we have economists and pundits generally watching the rate of change of many economic indicators. These people are visually interpreting a substantial slowing in the drop of many indicators as a sign prosperity is just around the corner. They are using extremely faulty rationalizations in their analysis.

In other words, if I showed you a time function on a two dimensional x-y chart, could you visually tell me what that plot would look like in the future? To the right of what was shown? Of course you couldn't. Not without understanding the function. In economics we clearly cannot define the function. That very statement is why Wall Street is blowing up. They thought they could mathematically reason the future. And the "the recession is over or nearly over" prognosticators are unknowingly espousing the same faulty idiocy used to curve-fit the quantitative models blowing up left and right on Wall Street. Someone takes ten or twenty years of data and fits the model to the faulty data. Even if fifty or seventy years of data is used it is completely faulty. And, there are no quantitative models used on Wall Street that I am aware of that use a data set anywhere near fifty or seventy years. And, even if they did, using seventy years of data is still a completely contrived and completely faulty analysis.

When I analyze the curve-fitted data, I see a shape that looks like Humpty Dumpty. So, therefore, I conclude he's going to fall off the wall. You know the rest of the story.

All the kings horses,

And all the kings men,

Couldn't put globalization back together again.

The world has permanently changed. Curve-fit that.