From a small northwestern observatory…

Finance and economics generally focused on real estate

Archive for December 2017

The WalMart Effect

leave a comment »

It’s a sleepy Saturday morning in the Pacific Northwest.  December 23.  Like so many Americans, I have some last minute shopping to do.  Like so many Americans, I’m dreading that, and looking around feverishly for an excuse to not do it.  Fortunately, I have a blog and a functioning computer.

Reading Reddit recently (and always looking for an excuse at alliteration) I happened upon an interesting map of the US showing the largest employer in each state.  I couldn’t help but notice that in a large number of states, WalMart was the largest single employer.  As someone who regularly travels the length and breadth of our great nation, I also couldn’t help but observe that many of the states where WalMart dominates employment are also at the lower end of the economic totem pole.  Aha!  I exclaimed, for I had suddenly stumbled on a distraction worthy of my Saturday morning.

Before I go any further, please let me stress that I have nothing against WalMart.  I’ve actually done a bit of work for them, in the long distant past (OK, maybe 15 years or so ago) on property disposal issues.  They have very tough standards for consultants, but they pay their bills (although, you earn the money!).  Indeed, I have very few bones to pick at all in this study, other than the obvious which is that political and civic leaders owe it to their states and communities to seek out exportive “basic employment” businesses.   Communities do not benefit by sending their money elsewhere (which is what they do when people shop at any retail establishment, as I will do later today).  They benefit when they send goods and services elsewhere in trade for money.  Note that “exportive” is a general term.  In Nevada, for example, the largest employer is MGM Resorts.  The customers are kind enough to haul large sums of money to Nevada from out-of-state and dump it in the middle of the street for the locals to scoop up by the bucketful.

My little study also does not account for states which have lots of exportive businesses, but no single entity which is dominant.  For example, Florida is a bit of an outlier, in that tourism is a big exportive business, yet WalMart is still the biggest single employer.  Go figure.  I would have guessed Disney.  In many states, the University system is the biggest employer, but recognize that Universities are huge exportive enterprises.  Money flows in from research grants, technology transfer, patent licenses, tuition from out-of-state and foreign students (the U.S. is still the preferred place in the world for an education), and alumni gifts, and all we send out are journal articles and educated individuals.  From a local perspective, higher education is the gift that keeps on giving.

Anyway, I downloaded the data.  There are 22 states in which WalMart (“WM”) is dominant, and 28 in which it is not.  I also went to the Census Bureau and grabbed the state-by-state median household income (“HHI”) stats.  I then ran three quick studies (and yes, it’s still morning).  First, a simple correlation shows that having WM as the state’s largest employer has a negative 51% correlation with HHI.  In other words, if your state has WM as its largest employer, your state is 51% likely to have lower-than-average HHI (yes, this is a simplistic way of looking at it, but bear with me).

I then ran a simple log-linear regression. (Since HHI is bounded at zero, the conditions for OLS don’t stand, so a log transform of HHI is needed.  It actually makes it a bit easier to interpret the data.  Plus, there are a few other reason, but I won’t bore you.)  The WalMart Effect, as I call it, is about 15.75%, with a very high (>99%) level of confidence.  In other words, if your state has WalMart as its largest employer, your state’s median household income is likely to be 15.75% below average.

Finally, I measured the actual numbers.  Of the 28 states which do NOT have Walmart as the largest employer, the average median household income was $62,116 in 2016.  If WalMart is the largest employer, that number was $52,290, for a difference of $9,818.  Thus, the Walmart effect….

Now, I’m going to go shopping.  I’ll eventually write this up in a little paper.  E-mail me and I’ll send it to you when and if I get around to it.

Written by johnkilpatrick

December 23, 2017 at 11:58 am

Posted in Economy, Finance

Tagged with ,

Livingston Survey strengthens

leave a comment »

One of my economic “touchstones” is the semi-annual Livingston Survey, begun in 1946 by the famed economist and journalist Joseph A. Livingston.  The survey continues today under the auspices of the Philadelphia FED.  Twice a year they survey a panel of economic forecasters on the key metrics of unemployment, GDP growth, inflation, T-Bill and Bond rates, and the S&P 500.  Not only are their opinions of interest, but also the change in the central tendency of those opinions over time.

For example, six months ago, the panel forecasted that year-end unemployment would be 4.3%, with a slight decline to 4.2% by mid-year, 2018.  Now, this forecast has shifted slightly downward, with an expected year-end unemployment rate of 4.1%, mid-year 2018 projected at 4.0%, and year-end 2018 at 3.9%.  These are decidedly low numbers, and suggest an econonomy at nearly full steam. (“Frictional” unemployment, which is the lowest level we would normally see, is generally thought to be close to 3%.)

Previously, year-end GDP growth was projected to come in at about 2.5%.  That’s now up to 2.9%, settling back to about 2.5% by mid-year 2018.  Projections of inflation are also solid, with CPI ending the year at about 2.1% and PPI (producer price index) at about 3.0%.  Both of these estimates are slightly lower than previously forecasted.  Intriguingly, CPI is forecasted to stay about the same in the coming year, while PPI should decline to about 2.0% by the end of the year next year.

The cost of debt is projected to increase in 2018, albeit at modest rates (and lower than previously projected).  Previously, the 10-year bond rate was forecasted to end the year at about 2.75%, but now should end the year at about 2.45%, according to the panel.  Rates should rise in 2018, but more slowly than previously projected, ending 2018 around 3.0%

Finally, the June survey projected that the S&P 500 would end the year at 2470, but now the panelists think the market will end the year at 2644.  (I note that the S&P sits at 2691 as I write this.)  The S&P is projected to end 2018 at 2805, or about 6% higher for the year.

The full survey also contains data on a variety of other topics (auto sales, corporate profits, average weekly earnings, etc.).  You can subscribe by visiting the Phily Fed at


Written by johnkilpatrick

December 18, 2017 at 11:25 am

So, folks, where are we going to live?

leave a comment »

Much has been said recently about housing starts being back up to where they were before the recession.  If this is the case, then why does Seattle, for example, have a 0.9 month supply of homes for sale?  As usual, the details are much more complicated than the headlines.

Prior to the “meltdown” (let’s say, 2004 – 2007), housing starts in the U.S. averaged about 1.865 million units per year.  Now, few analysts disagree that this was too many, but figuring out the right number is harder than one might think.  In 2008, the number dipped down to about 905,000, and hit a low of 583,000 in 2009.  Since then, the annual starts have trended up.  However, in 2016, we still were only at 1.207 million.  Of that, only 751,000 were single family units, compared to an average of 1.4 million single family homes per year in the 2004-2007 period.  Hence, nationally, we’re building about half as many homes as we were 10 years ago.

From 2004-2007, we started 7.462 million dwelling units in America, but in the past four years we’ve only started a total of about 4.432 million (all varieties).  That’s a shrinkage of about 3 million new homes, and most of that shrinkage is in the single family category.

One might posit that the decline in home ownership rates should have freed up some demand, and some of that’s true.  The home ownership rate in America peaked at 69% during the run-up to the recession, and dropped steadily after the melt-down, to a low of 62.9% in the 2nd quarter 2016.  As of the end of the 3rd quarter this year, it sits at 63.9%, or about 5 points below the peak of a decade ago.

There are about 76.146 million owner-occupied housing units in the U.S. today.  A five-percent swing in this number is a little over 3.5 million houses.  In short, we’ve now “absorbed” the decline in starts, and structurally we’re more-or-less “over” the recession, and we’re simply not building enough new homes to meet the demand.

Several consequences came out of the melt-down.  First, developing land takes quite a few years — five or more in the “hot” areas like Seattle, where land has to go thru a permitting and entitlement phase long before a house can be built.  All of this requires land planners, both in the private sector and downtown at the county or city hall. Many of these folks lost their jobs during the 2008-09 period, and indeed some county and city planning offices were eviscerated.  New home development frequently requires a significant outlay in public infrastructure, including schools, roads, and utilities.  Worse than that, many construction trades were gutted, with no replacements available.   Financing for acquisition, development, and construction is now problematic (although, arguably, it was too liberal pre-recession).

As such, it’s a sellers market for homes, and in hot markets, buyers compete by bidding up prices beyond reasonable levels.  Some pundits are nervous, and with good reason.

(Thanks to the U.S. Census Bureau for the October 31 data.)

Written by johnkilpatrick

December 11, 2017 at 11:29 am