The government's recent revisions to its key estimate of construction activity have proven so unnerving to economists that one important group today suspended its remodeling forecast for this quarter.

Harvard's Joint Center for Housing Studies announced today that it decided against issuing a new Leading Indicator of Remodeling Activity ( LIRA) because it needs to analyze Census Bureau revisions to the monthly construction spending report. That report, called the C-30 Construction Spending series, figures heavily into how LIRA is computed. The remodeling industry relies on the LIRA to predict how much Americans will spend on big-ticket home improvement projects over the coming 12 months.

In a blog posted today, JCHS research analyst Abbe Will said the revisions back to January 2011, included in May's construction spending report, boosted estimates of home improvement spending by 6% for 2011 but cut spending numbers by more than 10% for 2012.

"Typically, these annual revisions are minimal and, in the past, changes were always in the same direction as the original estimates, often revising the whole series downward somewhat," Will wrote. "This time, not only was the magnitude of the revisions significantly larger than in recent years, but the direction of the revisions was extremely divergent from what could have been expected based on previous annual revisions."

What's more, Will cited "worrisome" conflicts between the adjusted Census Bureau data and other economic reports, such as the National Association of Home Builders' Remodeling Market Index and the National Association of Realtors' Pending Home Sales Index.

"At this time, there is no obvious explanation for why the revisions to the C-30 improvements data were so extreme this year," Will wrote. "As part of the Joint Center’s investigation of this issue, we will be in contact with the federal agencies involved in collecting the survey data and developing these estimates to assess whether changes in survey methodology or weighting procedures, for example, might explain these large shifts." As a result, the next LIRA will be released Oct. 17, JCHS said.

Home improvement numbers actually aren't specified in the report. Rather, economists calculate their own home improvement number by taking the estimate for all residential construction and then subtracting the totals for new construction. A Census Bureau official told REMODELING that the home improvement numbers that go into the overall residential construction figure are based on a quarterly survey of consumer expenditures.

The official agreed with JCHS' contention that past revisions to the home improvement numbers were modest. She said Census Bureau workers are examining why the revisions occurred this time but as yet haven't identified any major factors that precipitated the changes.

Revisions are common to virtually all government economic statistics. At the bottom of the opening page of the construction spending report, the Census Bureau cautions readers that its residential construction estimates have a wide variation. For instance, while the Census Bureau believes residential construction in May rose by 1.2% from April, it also says the real change could be anywhere between a 0.1% drop and a 2.5% rise. "It may take two months to establish an underlying trend for total construction and as long as eight months for specific categories of construction," the notice adds.  Craig Webb is editor-in-chief of REMODELING. Follow him on Twitter at @craiglwebb or @RemodelingMag.