Apparently economists don't have a very good reading on the pulse of the new home construction market. For the second month straight, analysts predicted that housing starts would decline when they actually rose. In September they increased slightly to an annualized rate of 610,000, an rise of 0.3%, according to the Census Bureau. That doesn't sound great, but economists expected them to decline to 587,000. What should we make of today's report?

So here's the chart for housing starts, which are defined as the number of new homes where construction broke ground:

starts 2010-09.png

After rising by more than 10% in August, starts were essentially flat in September. But again, that's much better than the 3.6% decline economists expected. Last month's tally was also the largest number of starts since April -- before the home buyer credit expired. The still remain at a relatively depressed level, however.

While the housing starts data provides some lukewarm hope for construction jobs, data for new permits doesn't. The Census Bureau also reported today that new permits fell by 5.6% to 539,000 in September:

permits 2010-09.png

You can see that this is quite low. Other than the very low rate of 522,000 to 523,000 in March and April of last year, September's permits were the lowest in at least two decades. The number of new permits is more of a leading indicator than starts, so this implies that new home construction may fall again in the months to come.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.