Ladies and Gents: GDP has finally gotten its long awaited forecast

Today we will be finally creating our long awaited GDP forecast.  In order to create this forecast we have to combine both the forecast from our deterministic trend model and the forecast from our de-trended GDP model.

Our model for the trend is:

trendyx= 892.656210 + -30.365580*x  + 0.335586*x2

where x2=x^2
and the vector length we will make out to the 278th observation:

> x=c(1:278)

and our model for the cyclical de-trended series is from an AR(10) process:

GDP.fit<-arima(dt,order=c(10,0,0),include.mean=FALSE)

So lets say we want to predict GDP 21 periods into the future. Type in the following for the cyclical forecast:

> GDP.pred<-predict(GDP.fit,n.ahead=21)

Now when we produce our forecast we can't just add trendyx + GDP.pred$pred because the vector lengths won't match. To see this use the length() function:


> length(trendyx)
[1] 278
> length(GDP.pred$pred)
[1] 21

In order to fix this problem we are going to remove the first 258 observations from trendyx so that we only have 21 left:


> true.trend<-trendyx[-c(1:257)]
> length(true.trend)
[1] 21

Now we can plot away without any technical difficulties:

> plot(GDP,type="l",xlim=c(40,75),ylim=c(5000,18500),main="GDP Predictions")

> lines(GDP.pred$pred+true.trend,col="blue")
> lines((GDP.pred$pred+true.trend)-2*GDP.pred$se,col="red")
> lines((GDP.pred$pred+true.trend)+2*GDP.pred$se,col="red")

This code results in the following plot:

The blue line represents our point forecast and the red lines represent our 95% Confidence Level interval forecast.  I feel like the plot could be significantly cooler and therefore at its current appearance receives a 2 out of 10 for style.  It's bland, the x-axis doesn't have dates and there's not even any background color. If this plot had a name it would be doodoo.  A war must be fought against the army of lame plots.  Epic battles will proceed. Plots will be lost. Only one victor will stand.


Keep Dancin',

Steven J.

GDP Numbers: Actually Not That Bad

Gross Domestic Product (GDP), measures how fast or slow the economy is growing.  GDP reflects the final value of all output regardless of whether it was sold or placed in inventory.  The current report reveals that Real GDP (GDP adjusted for inflation) is slowing down from 3.7% growth in the last quarter to only 2.4% in the second quarter.  One thing to keep in mind is that this is the advance estimate so it is based on source data that are incomplete or subject to revision.  Lets just hope the revision isn't downward as 2.4% is already severely below most economists expectations.

Things included in the report:

Imports of goods are up 35.4% which is a sign that Americans are spending more (although this did put a damper on the GDP number).

Durable Goods only grew 7.5%- although this is not surprising given the severe cutback in the availability of consumer credit. Also one in four Americans has a credit rating of less than 600 which essentially prevents them from getting any credit for at least a while to come. 

Gross Private Domestic Investment- what businesses invest in plants plants and equipment (this number grew 28.8% which is healthy). This series tends to be ridiculously volatile but we'll look at it anyway.  Most of the increase has stemmed from nonresidential investment in equipment and software which saw a 21.9% increase and a surprising increase in residential investment of 27.9%.  Residential investment includes single family homes and apartment buildings.

Also the government came to the rescue somewhat with non-defense government consumption expenditures seeing a 13% increase.  This means that things are indeed getting built and that the government is responsible for putting some people to work.

This recession has been characterized by some (including Christina Romer doing a recent interview with the WSJ)  as a problem of deficient aggregate demand.
To get a better measure of pure demand within the U.S. economy look at gross domestic purchases under "Addenda".  This measure sums up total purchases by U.S. consumers and businesses regardless of whether the product was actually made in the U.S. or somewhere else.  This measure excludes exports and includes imports.  This measure is up 5.1% which is the largest increase in several years and definitely a positive sign.  Also an even better measure is final sales to domestic purchasers which was up 4.1%.  This measure excludes changes in inventories and exports to foreigners making this possibly the purest measure of demand.   So what have we learned? While the official measure of GDP is increasing at a decreasing rate the actual demand for goods by Americans has in fact been increasing at an increasing rate.   This is certainly better news than what most economist have you believe.   Furthermore, if final sales are increasing at a faster rate than GDP growth it is a sign of strong economic growth ahead as companies accelerate production to meet higher demand. 

Now for a look at deflationary risks:
Go to table 4 in the report.  Near the bottom of table 4 under implicit price deflator lies the deflator for gross domestic purchases.  Same as before this measure takes into account price changes from all purchases (including imports) and only saw an increase of 0.1%.  This is some scary disinflation that may soon turn negative to reveal deflation.

Another thing to look at is the Personal Consumption Expenditures deflator (or PCE deflator).  The Fed uses this measure and relies on it more than CPI because it is sensitive to ongoing changes in consumer spending patterns.  This measure is certainly showing disinflation 2.9% -> 2.7% -> 2.1% -> 0.1% and reveals that deflation is around the corner.

Research & Development: An Investment or Expense?

Apparently R&D is counted as an expense and not as an investment in the official calculation of gross domestic product (GDP).  This was first brought to my attention by the Levi Institute Multiplier Effect Blog.  One would think that research and development would be counted as investment which makes up a valuable component of GDP.  According to the Bureau of Economic Analysis (BEA) counting R&D as an investment would have increased GDP 2.7% from 1998 to 2007.  The new research from the Bureau of Economic Analysis concludes that:

"By treating R&D as investment, real GDP increased at an average annual rate of 3.0 percent over the period 1998-2007.  As in previous periods, growth in R&D investment continued to track business cycles.  R&D’s contribution to growth slowed in 2001 and 2002, recovered in 2003, and outpaced the expansion through 2007.  In 2002, business sector R&D subtracted from growth, but was more than offset by contributions from the government and nonprofit sectors."

R&D should be counted as an investment because it leads to innovation and growth.  R&D has been overlooked and should be treated more seriously as increased R&D spending leads to greater productivity growth.  It would also be valuable to track its role in business cycle fluctuations. 

GDP Numbers Revised Downward

From the Bureau of Economic Analysis:

"Real gross domestic product -- the output of goods and services produced by labor and property located in the United States -- increased at an annual rate of 2.7 percent in the first quarter of 2010, (that is, from the fourth quarter to the first quarter), according to the "third" estimate released by the Bureau of Economic Analysis. In the fourth quarter of 2009, real GDP increased 5.6 percent."

The 2.7% is a .3% revision downward from the "second" estimate and a .5% downward revision from the "advance" estimate.   

A look at the Advanced Retail Sales Numbers

The U.S. Census puts out a report each month called the Advance Monthly Sales For Retail Trade and Food Services. Why should we as economists care about what this report says? I generally try to care about any report that the Fed deems important. So why does the Fed look at retail sales??? Retail Sales is used to compute Personal Consumption Expenditures, which is the most important component of calculating the nation's GDP. Changes in real GDP correlate well with changes in real retail sales.
One thing to keep in mind is that retail sales is only measured in nominal terms, which means that no adjustment is made for inflation. This makes it difficult to tell whether a jump in the numbers came from consumers actually purchasing more or paying more to cover higher prices charged by retailers. Also the initial retail sales releases tend to be extremely volatile and thus misleading. This is partially because the advanced estimates are based on a relatively small sampling size. A three month moving average of the data provides a more accurate picture of whats occurring in the economy. For a look at the percentage changes go to table 2A which shows that retail sales dropped 1.4% in May but is still up 7.4% from last May. If you exclude motor vehicle sales and auto-related products (which tend to be extremely volatile) we see that the fall in the retail sales numbers is by a smaller 1.1% which (according to this number) is still up 6.1% from last May. So what exactly are U.S. consumers purchasing? Apparently gasoline which is up in usage 20.2% since last May (although this increase may represent an rise in gasoline prices). Furthermore, we see that the end of the April 30th homebuyer tax-credit brought about a drop in demand for building materials( -9.3% drop from last month). Overall the report is generally a disappointment among the economics community (check out WSJ's Economists React: 'Weaker Underlying Picture Revealed' for Consumer) as it suggests we have a struggling consumer. The Census Bureau should track polo hat sales: If demand for ridiculous items picks up then the "true American consumer" will be perched back up on their throne.