The economy is forecast as weak in October 2011 – but in perspective should be roughly similar to the September economy.
Although recession risks exist and Econintersect has pointed out several data points (imports and equity markets) that are recessionary, the preponderance of forward looking data points are weak but not recessionary.
Overall, the Econintersect Economic Index has a less good 3-month moving average near contraction – but the last 2 months have shown marginally improving data. It also should be highlighted that ECRI – a proven business cycle forecasting service - has called a recession. In a statement to clients early last week, it stated:
…… that the U.S. economy is indeed tipping into a new recession. And there’s nothing that policy makers can do to head it off.
Econintersect’s methodology only views one month ahead while ECRI says its indicators look at least 6 months ahead. ECRI is claiming a recession is a done deal (and Econintersect is a believer in the methodology). However, our indicators have not yet entered recession territory.
The major coincident data points (all of which is at least two months older than this forecast period) on average remain weakly in expansion territory. The trend lines of this data are generally flat or trending down. Econintersect uses the coincident data trend lines to validate its economic model.
Honestly, we are afraid that the New Normal economy has different dynamics than any economic model.
By back testing its model, Econintersect has determined its index was not giving a proper 30-day forecast. On the other hand, what was labeled & published as the transport segment of the Econintersect Economic Index (EEI) has been providing a fairly accurate real time outlook. The former transport segment is now the Econintersect Economic Index. Please see the end of this post for a discussion of the “new” EEI.
The question remains whether the USA economy is entering a recession.
Last year Econintersect reacted to this poor data and the rapidly degrading trend lines calling a “recession watch” – only to have the data to improve. This year, the contraction appears more severe in the EEI – but there are two months now of improving data (although the 3 month removing average continues less good). The other factor to be noted is that the EEI looked much worse last year in real time because it contained data that has since been improved by backward revision. The “old” EEI actually went negative for three months in real time, but after revisions this year stayed in positive territory.
The economic state is currently close to contraction territory. It is weak. And any economic kick in the stomach could result in a true contraction. Econintersect warns that most elements of the economy have NOT recovered to pre-2007 recession levels. If this was prior to WWII, economists would consider the current USA economic state as a depression.
Economic Forecast Data
Econintersect’s Economic Index (EEI) is designed to spot Main Street and business economic turning points. We began warning in April of an economic slowdown. In this October forecast, we are continuing to advise that an economic cycle expansion MAY be beginning. It will take several months of data to confirm IF this is a turning point.
This forecast is based on Econintersect’s non-monetary based economic index, which counts “things” that have shown to be indicative of direction of the economy 30 days in the future. Note that the Econintersect Economic Index is not constructed to mimic GDP (although there are general correlations), but tries to model the economy seen by business and Main Street.
The EEI is not a monetary based index. It does not need to adjust data for inflation.
While continuing to expand in some areas, the economy is contracting in other areas. Where contraction is occurring, it is not severe in any one of our pulse points. And any expansion likewise is mediocre at best.
(Click charts to expand)
The red line on the EEI is the 3-month moving average, which is at .05, while the monthly index is at 0.20. A reading of 0.20 is not good.
It takes several months (or quarters) of negative data before a recession is determined. Note this index does not measure severity of contraction – only that a contraction is occurring across weighted pulse points.
Jobs Growth, While Not Good, Is Not Recessionary Either
The NBER recession dating committee used non-farm payrolls as the marker for the 2007 recession.
Since our last forecast, the August 2011 BLS jobs report showed zero growth – a recession marker. However, the BLS jobs report historically is inaccurate in real time, and is not accurate until several months following the initial data release. Further, the ADP employment data showed a continuing less good (but positive) employment picture.
Econintersect’s Employment Index is based on economic elements that create jobs. Econintersect’s Jobs index (explanation here) was at a cycle low in June. The jobs index measures the historical dynamics that lead to the creation of jobs, but it is not precise as many factors influence the exact timing of hiring. This index should be thought of as a measurement of jobs creation pressures.
At this point, the jobs growth year-over-year (YoY) is well above Econintersect’s Index. For July 2011, Econintersect forecast private non-farm employment growth at 135,000, while the actual number came in at 154,000. In August 2011, the index predicted employment growth of 145,000 – and the actual number came in at 17,000. The index projects both September and October 2011 non-farm private jobs growth at 145,000.
As you can see from the red line in the graph above, Econintersect is projecting deteriorating jobs growth for the next six months.
The Econintersect Economic Indicator© (EEI) Redocumented
The EEI was originally documented in the October 2010 forecast. As this is a “revised” EEI, this section re-documents this index. In a mechanical sense of the change, Econintersect has removed the portion of the index that was meant to “quiet” or remove noise. This deleted portion of the index was subject to significant backward revision from the data sources (limiting its effectiveness in real time), and in execution was not allowing the index to “lead.”
Consider that most economic indicators (data) undergo backward revisions – and although they are eventually correct, they are not correct when first issued. The EEI is almost entirely made up of fixed data points – and subject to little backward revision.
There is no change to the elements or methodology of the remaining index, which was performing as advertised. It was this new EEI that allowed Econintersect to declare a cycle turning point in April 2011.
The EEI is built using mostly non-monetary pulse points below retail sales / personal income receipt. Further its methodology handles seasonality by considering rate of change of year-over-year data.
The EEI monthly index went negative beginning in November 2007, but the three- month moving average went negative in April 2008 (versus NBER December 2007). This is not surprising as this is a Main Street index – not designed to mimic GDP. One major difference between GDP and the EEI is the turning points. Thus it should not be surprising that the EEI did not see the Great Recession contraction ending on Main Street until April 2010 (versus the NBER June 2009).
Data for several elements of the EEI were not available prior to 2000. It is comprised of 4 data groupings which are weighted. This is designed to be a New Normal index, and each element has a historical high correlation to leading economic turning points.
- government spending
- industrial production
- transport (measures of consumer and business demand)
- export / import (measures of consumer and business demand, as well as a measure of global demand)
Note this index is not trying to quantify the strengths of expansions or contractions – only that a contraction is occurring across weighted pulse points. It is a relative index to the preceding month – weighting and counting the elements that are better or worse.
In the event Econintersect believes the chance of a recession is highly possible, a recession watch will be issued. If that trend continues, the watch will be replaced by a warning. If contraction becomes endemic, a recession will be called.
Econintersect’s methodology has been borrowed from the alert system used by the U.S. Weather Bureau:
Watch: A recession is possible.
Warning: A recession is probable.
Recession: A recession has started.
Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.