How – and why – the Triangle fares in new Moody’s HQ2 rankings
Note: feel free to comment on this article via this LinkedIn post.
Another week, another Amazon “HQ2” ranking by a respected company. And, another opportunity for the regional business community to highlight the Triangle market’s competitive advantage in mobility – and to correct or at least clarify what the new set of rankings mean and do not mean.
A recent model by Moody’s Analytics compares dozens of metro areas against five broad categories listed in Amazon’s “HQ2” RFP – business environment, human capital, cost, quality of life, and transportation – and incorporates a bonus “geography” category.
When Moody’s considers the first five categories, the top 5 metro areas are Austin, Atlanta, Philadelphia, Rochester, and Philadelphia. When Moody’s geography category is incorporated, the top 5 list ranks Philadelphia first, followed by Pittsburgh, New York, Atlanta, and Rochester.
The Triangle (“Raleigh”) does not make the top 10 in Moody’s analysis, whether we include 5 or 6 categories, given the selected factors for the various categories and standard weightings. And once again, transportation is our worst-scoring area according to the model – despite the many competitive mobility advantages of the Triangle market.
When all is said and done, we end up ranked 13 nationally, with a middling raw score of 2.84 out of a maximum of 5 points, given the five core categories. If geography is included, our score drops slightly to 2.82, but we move up to 11, just outside the top 10 rankings.
Of course, what you score, and how you score, largely dictates how well you score.
Here, I review the Moody’s analysis and offer a more in depth perspective of several factors contributing the Triangle region’s transportation and overall rankings. All factor scores below have a maximum value of 5 in Moody’s comparative ranking system.
The Triangle (Raleigh) rankings by category
- We scored well in the “business environment” category in all aspects, except for state incentives, giving us an overall score of 4.2.
- Our “human capital” score was a moderate 2.5, with our comparatively lower population—partially caused by the exclusion of the Durham-Chapel Hill metro area from the model calculations—creating a challenge.
- Our “cost” score was a little better at 3.1, due to moderate housing prices and tech job salaries combined with an excellent overall cost of doing business.
- Our “quality of life” score, which combined information about crime, arts and entertainment, restaurants, schools, and diversity, was fairly average at 2.7.
- In terms of Moody’s bonus “geography” category, we again scored in the middle at 2.7. This factor gives points for a northeast/mid-Atlantic location (and somewhat lesser points for a southeast or midwest location), and also assigns points based on proximity to one of Jeff Bezos’ homes as well as the current number of Amazon fulfillment centers.
I have saved our “worst” category – at least according to Moody’s – for last. Our transportation score, which incorporates the percent of commuters who take transit, walk, or bike, plus our travel time by car and transit, and the total number of air travelers, was our lowest score of the five categories only 1.7 out of a maximum 5.
How can it be that a region with some of the lowest levels of traffic congestion for growing major metropolitan areas in America, served by the most connected medium-sized airport in the country, scores so poorly in mobility?
Let’s take a closer look at the elements in the transportation category.
- Our driving commuting time is reasonable, and we scored well, which is not surprising.
- The more people who actually commuted by transit, walking, or bicycling, the higher your score. Our spread-out metro area does not lend itself to good performance in any of these areas, and our low scores reflect that.
- The use of public transit commuting time, while understandable conceptually, leads to ridiculous scoring outcomes – in this case, that benefit us. We actually score better than New York, Boston, Chicago, and Washington for this factor. This cannot be because our system is currently so much more travel time efficient than those extensive transit systems. More likely, it is the opposite: our transit network is so small that you just can’t physically travel that far by bus, which likely caps the average or maximum travel time via transit.
- Air enplanements is fair and a reasonable proxy for air travel options, and we did fine given our market size.
The default Moody’s model assigns an equal weight to the individual elements in transportation and indeed all categories. Since there are separate factors for both pedestrians and bicycles, and two different transit factors, but only one commuting factor and one air service factor, this means that the model effectively treats both nonmotorized transportation (i.e., walking and bicycling) nd public transportation as twice as important as either car commuting or air travel.
I would argue that walking and biking are important, and transit is important, but not twice as important as travel by car or via air, particularly given the proportions of people using the various modes, in most markets.
An alternative weighting of transportation
How would our score change if we were to adjust some of the weights of various transportation elements?
Well, if we were to focus on just the commuting elements, and eliminate the pedestrian and bicycle elements as well as the two transit elements, and make no other changes, we would end up ranked 6th overall. In fact, if we were to also incorporate air service, our ranking would still be 6th.
If we were to decide to incorporate public transit, and just use the transit travel time factor (which benefits us) we would end up in the top 5 nationally, at 4th overall.
As noted above, what you score, and how you score, dictates how well you score. In our case, the core transportation realities and mobility future of the Triangle region cannot be easily captured by traditional scoring methods that focus on broad input data of current usage patterns which are, in effect, simply a static look in the rear-view mirror.
None of this is a criticism of Moody’s model. It is an acknowledgement of the limitations of any standardized evaluation which by necessity must use past performance data in a number of categories to evaluate potential metro area future success.
My sense is that Moody’s offers a very useful model that incorporates the key HQ2 categories, includes a very helpful geography element, gives the user the flexibility to modify the weights and review the individual factors, and acknowledges the limitation of using average or identical weights.
However, an inherent challenge – and not just with Moody’s model – is that most news articles just take a model’s calculations, assumptions, and weightings as written. While this provides a simple snapshot, it also creates a vacuum of understanding, and fortunately it provides an opportunity to offer context.
The bottom line is that our market, which is uniquely positioned and connected for car commuting regionwide today, and which has established a sustainable, accelerated funding framework with a flexible approach for future multimodal transportation success, cannot be easily described by traditional methods of evaluating regional mobility.
We thank Moody’s for creating a logical, flexible, and accessible model. And, we wish Amazon well, wherever they choose to invest – by whatever criteria they select.