Village Solutions

Retail Density: Apples to Oranges

Over the past decades I have worked in almost every major metropolitan area in the United States.  In this capacity I often encounter reports that the subject market is significantly under stored.  While doing my market research I am always amazed to find newspaper articles reporting that a market has significantly less retail space than the national average which always turns out to be wrong.   So, I will share some insights into one of the biggest myths in the retail real estate industry – “my market has significantly less retail space than the national average”.

The proliferation of this myth falls right into the lap of an uninformed media which appears to want to take a position of supporting growth or showing that an area is over stored with retail space.  This is because of the rapid expansion of power centers in the 1990s; lifestyle centers after 2000; and free standing mega stores in recent years have all contributed to a crowded retail landscape.  Therefore, in an attempt by the media to quantify the amount of retail space and to compare the density of retail in one market to another, retail space per capita has been used as a common indicator.  However, our analysis concludes that these types of numbers are among the most misquoted and misunderstood data points in the analysis of retail real estate.  The problem is like the old joke – “one lies and the other swears to it”.  In this case, retail space per capita is the number that is often, if not almost always, reported inaccurately and then the next article repeats it all over again and before you know it, you have a fact that everyone relies on.  But, when comparing vastly different markets and using so called national averages one may end up with nothing more than a comparison of apples to oranges.

The most commonly used database on retail space comes from The National Research Bureau (NRB); however, they only include shopping centers and not the total of all retail space. Consequently, shopping center data has often been incorrectly used as a total for all retail space and then compared to the local population to generate a per capita number.  In brief, the retail space per capita that is often quoted is actually shopping center space and not retail space.

This type of comparison works well as a barometer across the United States, but it is highly inconsistent when using it as a source of retail density in major urban areas that have a large amount of retail not located in shopping centers.

According to many, NRB[1] is the premier provider of retail real estate information in the U.S.  Its database of information contains information on over 40,500 shopping centers which is the most comprehensive and detailed information source on U.S. retail properties available. Likewise, NRB has prepared the Shopping Center Census for the past 20 years, which is published by the International Council of Shopping Centers (ICSC) and by the U.S. Department of Commerce / U.S. Census Bureau in Statistical Abstracts of the United States.  The NRB census is widely considered the authoritative source to calculate the retail area per capita.  While the NRB is an excellent source for shopping center gross leasable areas (GLA), it is insufficient when analyzing per capita GLA for urban areas for a number of reasons.

The retail GLA per capita of the United States is estimated to be in the 22 to 25 square feet of retail space for every man, woman and child. However as mentioned previously, the data only includes shopping centers above a minimum threshold.  In this context a shopping center is defined as single develop which has been planned and built as a dedicated unit with its own parking.  Conversely, the all free standing stores, small shopping centers and urban sidewalk facing retail is generally excluded in the NRB statistics.  Therefore, in highly concentrated urban areas where a large portion of the retail is free standing, the GLA per capita is significantly understated. For example, New York City has one of the lowest GLA per capita rates in the nation at approximately 8.

Another source used to cite total GLA is the Retail Tenant Directory.  Although this includes free standing retail, it only considers the 5,400 largest chains[2].  Therefore, it also leaves out a large amount of retail which skews the GLA per capita.  Therefore, when total “Retail Space” is calculated the numbers jump dramatically.  As an example the total US Retail Space per capita in 2008 was 41.5.  Similarly, when you consider the total retail space in major metro areas (shopping centers and free standing stores) the real per capita calculations for retail space is much more consistent with the national average.

For example, as of 2003, the Miami-Fort Lauderdale-Miami Beach region had 42.9 sq ft per capital of total retail vs. a national average 39.8[3].  Based on this information, Miami’s Retail Space exceeds that of the national average.  Therefore, retail GLA per capita grossly misrepresents the market since most reports cite the NRB as their source which only includes shopping center retail.

It should be noted that Florida is widely considered to have one of the highest GLAs per capita in the nation with Fort Meyers, Fort Walton Beach, Fort Lauderdale, West Palm, Orlando and Melbourne all  being ranked in the top 25 and having a GLA per capita in excess of 25.  But again you cannot take these numbers at face value.  As we all know Florida is one of the fastest growing states in the US.  Outside of Miami, the largest percentage of retail in the state is located in strip centers built over the last 35 years and not in downtown’s.

To get a better handle on the subject, we spoke with ICSC who confirmed that the retail GLA per capita for the U.S. is 20, but it only takes into account shopping centers. As stated early this massively understates the retail GLA per capita of urban areas where the majority of retail is NOT in shopping centers. For instance, as outlined above New York City’s reported GLA of 8 per capital, implies that it is one of the most underdeveloped markets for retail in the U.S. The researcher at the ICSC also confirmed that Miami doesn’t pick up the majority of their retail for the same reason. The point is that while retail GLA is a good starting point for most markets in the U.S., it is NOT reliable for the major urban markets.

To prove the above points, let’s look at two recent studies in recent years.  In 2003 CoStar calculated the total retail space per capita (shopping centers and everything else) for the top 50 U.S. markets. Those 50 markets had an estimated average of 43.71 square feet of retail space. Portland had the third lowest retail space per capita at 27.95 square feet, trailing Long Island and Charlotte. The market with the most retail space per capita was Southwest Florida at 74 square feet, followed by Richmond, Winston-Salem, Greenville, Tulsa, Oklahoma City, Toledo, San Antonio, Jacksonville, and Birmingham.

---

[1] National Research Bureau   http://www.costar.com/

[2] “The Retail Tenant Directory is a publication containing in-depth profiles of over 5,400 retail chains across the U.S. and Canada.”    http://www.retailtenants.com/faq.shtml

[3] Information provided by ICSC Research Department based upon our request.  They complete national research about every year, but only research major metro areas every several years.

Tags: Florida, GLA, ICSC, leasing, lifestyle center, Miami, NRB, real estate myths, retail space per capita, square footage

This entry was posted on Monday, July 27th, 2009 at 10:47 am and is filed under Real Estate, Real Estate Development, Retail, Shopping Centers, Specialty Retail. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.