Scott Walker still misleading on the Philly Fed

Look what Walker tweeted October 3, 2013 3:14 pm:
philly tweet walker

“Leading”? But I thought the Philly Fed said these numbers were not to be used to rank states one to the other.
So the numbers show Wisconsin leading compared to ____?
Poland? Mexico? Mars?

You may have noticed that Philly Fed called Walker on earlier ranking talk. This was
published September 28, 2013 6:00 am in Philly Fed says Scott Walker, WMC offer misleading data on Wisconsin economy:

philly to cap times

Here’s the actual “Leading” we’re doing.

leading states

If you look at the map, you’ll see that Wisconsin is in a pack of states that the Philly Fed anticipates will have a state coincident index change of 1.5-4.5% in the next 6 months.

 

Here’s the color key for the states:

color key philly map

And we can see from the title that these are “Leading Indexes”.

leading indexes

Well, it’s not bad news, but it’s also a far cry from “We’re number 1”!

Scott Walker’s version of ‘leading’ is always going to involve a foam #1 and the twitter equivalent of twerking.

That’s why it’s very helpful to have good reporters and editors to help interpret what’s going on AND why I was disheartened to see this tweet from reporter Matthew DeFour:

defour

I asked Matt a couple of questions about that on twitter including where he got the 30 year data. Hopefully he’ll provide some answers. Notice that Robin Vos retweeted DeFour’s tweet (probably while smirking).

If you’d like to better understand why the Philly Fed called Walker on his ‘misleading’ claim earlier, here’s the text of the email Paul Flora sent to Mike Ivey which I have copied and pasted from the Capital Times:

From: Flora, Paul [mailto:paul.flora@phil.frb.org]
Sent: Friday, September 27, 2013 4:00 PM
To: Mike Ivey
Cc: Puenpatom, Tosmai; Elliott, Thomas
Subject: Response on the Coincident and Leading Indexes

Mike,

I am responding to your query to Tosmai Puenpatom about our coincident and leading indexes. I head up our team of economic analysts, including Tosmai who is currently our lead researcher on the indexes. We have prepared the following explanation regarding the use of state rankings based on our indexes.

We do not consider state rankings based on the coincident and leading indexes to be valid.

* The size and maturity of state economies influence the relative size of economic change within a state; an older, mature economy, such as New York, tends to experience smaller percentage changes in its trend than a smaller, younger economy, such as North Dakota. Comparisons between the two are not very meaningful.

* As a proxy for state GDP, the coincident index measures some states better than others. For example, the index components are heavily influenced by employment rather than output measures. Agricultural activity is largely excluded from the employment statistics and mining activity is heavily influenced by prices. In Pennsylvania, shale gas production has continued to grow, while mining employment has been edging down since January 2011. Some of the unique characteristics of each state’s economy are missing from this approach.

* Our research suggests that annual benchmark revisions of nonfarm payroll employment and the unemployment rate contribute to revisions in the coincident index. Benchmark revisions may be random across states but may also be positively or negatively correlated. One group of states may be revised upward for a common reason, while other states may be revised downward so that the national aggregate remains the same. Indexes based on data since the last benchmark base month must be used with caution.

* There is significant volatility in components underlying the indexes, especially for nonfarm payroll employment. This monthly volatility generates substantial volatility of the indexes for each state. Since the volatility of the percent change in these indexes is relatively large compared to the average change, an individual state’s ranking based on the percent change can jump wildly from one end of a relatively narrow range to the other. Rank order is not persistent, thus state rankings are misleading.

However, the state coincident indexes track an individual state’s GDP quite well, and in that sense, they are a useful proxy for state GDP. Their principal advantages include being more timely and more frequent than state GDP estimates.

Regards,
Paul

Paul R. Flora, Senior Economic Analyst,
Research and Policy Support Manager
Research Department
Federal Reserve Bank of Philadelphia
Ten Independence Mall
Philadelphia, PA 19106-1574

215-574-6649
215-574-4303 FAX

SOURCE

 

FOOTNOTE:

More on what the Philly Fed’s State Coincident Index is.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s