The Cycle of Development, Optimal Redevelopment, Redevelopment Goals and Benefits, and Barriers to Redevelopment

Before I review the foundations of real estate development finance and the role of public-private partnerships in redeveloping the United States, I need to show where P3s fit in the development/redevelopment cycle. I start by describing the cycle of urban development and what I call "efficient redevelopment." This is followed by a review of impediments to efficient development, and I conclude with the role of P3s to facilitate efficient redevelopment.

The Cycle of Development

That urban areas transform themselves over time is certain. Miles Colean (1953) calls this the "cycle of development." Larry S. Bourne (1967) provides a succinct review of the process, which comprises an initial period of construction followed by a period of increasing value and function, then a period of increasing maintenance costs and deterioration, perhaps leading to idling or abandonment, and then a period of redevelopment as the old structures are replaced.

Consider the normal life of a building. It is built initially to serve an investment horizon and becomes obsolete either because of economic factors (where the building is more expensive to maintain than justified by revenue streams) or functionality (where markets have changed, leaving the building unsuitable for its initial use) or both. As the structure loses value through a process called depreciation, the land on which it sits will normally gain value, especially if the urban area is growing.

Figure 1.1 illustrates this appreciation in land and depreciation in the structure assuming a new building is built with a fifty-year useful life, which is common for one- and two-floor office buildings. When its doors open, the building accounts for 80 percent of the property value and the land for the remaining 20 percent; this is a typical building-to-land relationship for commercial buildings. The building depreciates over a fifty-year period, or 1.6 percent annually, and becomes worthless (except for any scrap value) in the fiftieth year. Land, on the other hand, gains value at about the rate of growth of the urban area, compounded. If the population or employment growth rate is 2 percent annually, the land value increases at this rate (net of inflation). By the twenty-eighth year, the land is worth more than the building. Some years before and after this happens, the investors reassess their investment and, ideally, renew the site by replacing the initial structure with one consistent with the highest and best use over a new investment horizon.

For the most part, nonresidential space is not durable. Overall, the United States has about 100 billion square feet of enclosed space used for such nonresidential purposes as retail, offices, institutions, and so forth. About 70 percent of all nonresidential space is housed in buildings of one or two floors. In any given year, about 2.5 billion square feet of nonresidential space becomes idled or is replaced—2.5 percent annually.

In contrast, residential structures are quite durable. The United States has about 130 million residential units, but only 500,000 residential units—about 0.5 percent—become vacant or are replaced each year. I have estimated that the typical residential unit lasts about 170 years (Nelson 2004, 2013a). Pitkin and Myers (2008) estimate that units last 200 to 500 years. Whatever the length, planners and public officials need to understand that residential development is very durable, not because the structures themselves are built to last a long time, but because occupants will maintain the unit through repairs and rehabilitation for decades or even centuries. Most nonresidential development, in contrast, is not durable and needs to be replaced about every 20 to 40 years.

Urbanized land thus goes through a series of changes over decades and centuries. The first building on a site, for instance, might be a neighborhood grocery store. As the building ages, it becomes more expensive to maintain, so profit (revenue net of costs) goes down. In the meantime, the land value goes up. The "opportunity" cost of keeping the land in its current use goes up as profit in the current use goes down. At some point, the landowner incurs the cost of demolition and rebuilding to increase profits by going to the "highest and best" use of the land. Maybe the new structure is a low-rise retail store. In a few more decades, the next highest and best use might be a midrise office building. In theory, redevelopment of the built environment would be seamless, leading to ever higher and better uses over time, as illustrated in figure 1.2. In practice, this is rarely the case, for reasons I outline next.

Optimal Redevelopment

The first buildings to be constructed in an area are often small and built of material that is easy to dismantle. At some point, buildings become of such size and durability that they may be difficult to replace, especially if market conditions do not warrant the expense of both dismantling and rebuilding the site. The result can be what Bourne (1967) calls a "constrained" process of redevelopment. This could lead to blight as the structure becomes idled or vacant and its presence discourages reinvestment in the area, thereby delaying redevelopment beyond that which is "optimal."

In a classic paper theorizing the optimal timing of redevelopment, Donald Shoup (1970, 43) demonstrated that the optimal time for redevelopment of urban land depends on four factors:

The optimal date for development or redevelopment of urban land depends on (1) the discount rate applying in the real estate market, (2) the property tax rate, (3) the earnings in any interim use, and (4) the way in which the highest and best use of the land is expected to change in the future.

I refine Shoup's principles for application in this book. The first is the "discount rate" that is applicable to the local real estate market. Put simply, this is the rate at which future revenues net of costs ("profit") are discounted to the present to allow for a fair comparison of alternative investment choices. A high discount rate means the investor is willing to pay less for something, presumably because risks are higher.

The discount rate is akin to the capitalization rate, or "cap rate," which is the ratio between the net operating income (NOI) of a real estate investment (rental income less operating expenses) and its market value: NOI/value. If the NOI is $100,000 and the building has a value of $1 million, the cap rate is 0.10 ($100,000/$1 million). If the building has a value of $2 million, the cap rate is 0.05. Generally, the higher the capitalization rate is, the sooner redevelopment will occur. This is because the higher the cap rate, the lower the value of real estate, and thus the more attractive it is for redevelopment. (I discuss capitalization rate mechanics in the "Real Estate Finance Concepts" section in chapter 3.)

Generally, the market determines the cap rate, so the key variable in estimating value is the NOI. If the NOI goes down, perhaps because taxes go up or rents go down (as the building deteriorates with age), the building value goes down. As value declines but the local market is stable or growing, the optimal time of redevelopment will occur sooner rather than later.

Related to the NOI is the second redevelopment factor: the level of property taxes. Property taxes are considered an expense, so the higher the taxes, the lower the NOI, and therefore the lower the value and the more likely a site may be redeveloped. Hence, one outcome of high property taxes is accelerated redevelopment, whereas a lower rate may defer it.

The optimal use of land will be affected by Shoup's third factor: the earnings in any interim use preceding the next highest and best use. If there is little or no income, the optimal timing of development will be accelerated, but if there is enough revenue from an interim use, optimal redevelopment will be delayed. Using the case of downtown surface parking lots, I will demonstrate how inefficient property taxes can increase earnings from interim uses of land, thereby raising the NOI and deferring the optimal timing of redevelopment.

Property taxes are an important part of real estate investment decisions. Nationally, property taxes average about 1 percent of the market value of property, though there is wide variation among states and local governments. One percent on a $1 million commercial property would be $10,000 annually. If the NOI is $100,000, the property tax would be equivalent to 10 percent of net revenue.

Property taxes are based on the value of property, but "value" can mean many things. The value of a home that sells for $100,000 would be considered $100,000, and it would be taxed accordingly. If a home does not sell for decades, its value would be estimated, and ideally that value would be equivalent to what it would sell for and it would be taxed accordingly. Although this seems straightforward, its application to income-producing property gets complex.

Suppose there is an acre of vacant land in the middle of downtown. Its property taxes would be based on its estimated sales price. If nearby lots sold for $1 per square foot, this acre would be worth perhaps $4,356,000 and it would be taxed accordingly. A 1 percent tax rate means the owner would need to pay $43,560 annually in property taxes.

The owner may wisely convert this vacant piece of land into a surface parking lot. Capital costs are mostly putting asphalt over the land and building an attendant's shack. Operating costs are just the attendants and their benefits, a business license, modest grounds keeping, and property taxes. But instead of paying $43,560 in taxes, the local property tax assessor values the parking lot based on its current use ("use-value") and not the market value of the property. To calculate the use-value of the parking lot, the assessor applies the capitalization rate to the parking lot NOI. The example of this appears in the top half of table 1.1. The use-value would be $278,421, or about 6 percent of the market value of land. At 1 percent, the property taxes would be $2,784, not $43,560. Thanks to property tax policy, the interim use generates sufficient cash flow to carry the property.

But what if the policy defers redevelopment beyond that which is optimal? The property owners have little incentive to sell and all the incentive to wait until the market value of their property rises to such a high level that they become willing to sell. In the meantime, the development that would have occurred but for the property tax policy does not occur or, worse, is deflected elsewhere into places where such development is suboptimal. Development patterns would thus be skewed. The property owners themselves could be considered speculators, as their holding costs are artificially low, essentially being subsidized by other payers of property tax. When they sell, they would receive speculative as opposed to normal gains (Shoup 1970, 44).

In contrast, suppose this parking lot were assessed taxes based on its market value as opposed to its use-value. This situation is shown in table 1.1. The owners would incur a loss of $15,718. Unless they are willing to pay those losses year after year, they may be induced into developing or selling to a developer. Development would thus be optimal. A few cities have moved from use-value to market-value property tax systems in their downtowns, with desired effects (see Oates and Schwab 1996).

The last factor relates to expectations of what the highest and best use of property may be in the future. An owner of property may choose to keep it undeveloped (or underdeveloped as an interim use) until the market favors a much more intensely developed project—perhaps a high-rise tower surrounded by lower-rise offices and residential buildings. During the U.S. suburbanization process, for instance, some landowners decided to wait until low-density residential development surrounded their parcel and would then build a shopping center serving those new households. The owners may risk waiting too long to develop, however, as markets change over time.

Planning Goals for and the Benefits of Redevelopment

In my view, redevelopment should be based on meeting these five planning goals (see Nelson and Duncan 1995), which are described further in the following sections:

• Maximizing environmental quality

• Minimizing the cost of publicly provided facilities and services

• Maximizing land-use interactions

• Fairly distributing the benefits of development

• Elevating the quality of life

MAXIMIZING ENVIRONMENTAL QUALITY

Maximizing environmental quality does not necessarily mean development must have no adverse environmental outcomes. What it does mean is maximizing it within the context of meeting other societal needs, such as real estate development. Two key environmental benefits associated with redeveloping existing places rather than developing greenfields are (1) preserving open space so as to continue receiving ecosystem service benefits and (2) reducing carbon emissions so as to improve air quality and reduce the pace of climate change.

In my book Reshaping Metropolitan America (Nelson 2013b), I estimated that there are enough parking lots to support all new and redeveloped nonresidential needs and all multifamily development needs for the United States; in some fast-growing metropolitan areas, however, this is not the case. I also estimated that if all new residential and nonresidential development occurred on greenfields, roughly 15 million acres of land would be consumed. Using analysis reported by Mertens and Rubinchik (2006), I estimated that the present value cost in lost ecosystem service benefits would be about $4 trillion over the next century. If all development occurred on parking lots, this figure would be much lower, arguably even zero.

Moreover, if all new development occurs on existing developed land, two things happen to the consumption of fossil fuels. First, miles traveled per vehicle may go down because the outward spread of urbanization is halted. Second, new development on existing developed parcels can reduce the distances between origin and destination. As the number of vehicle miles traveled (VMT) is reduced, so are the greenhouse gases that science has found to influence global climate change (see Ewing et al. 2008). I estimated that, when infill and redevelopment projects are coordinated, the cumulative and synergistic air emissions reductions range up to 40 percent without transit options and up to 50 percent with them.

MINIMIZING THE COST OF PUBLICLY PROVIDED FACILITIES AND SERVICES

An extensive literature shows that the costs of delivering a large range of public facilities and services vary by the location, density, and configuration or mix of development (see Nelson, Bowles, et al. 2008; Nelson 2013b). The same number of users, for instance, can be served by a ten-inch water main extending one mile or ten miles, but the cost to serve those users over ten miles is literally ten times more expensive than serving the users over one mile. This would not be an issue if users paid fees or taxes based on the actual cost to extend facilities to their properties. I call this "full cost" charges. Unfortunately, most locally provided facilities are financed based on "average" cost charges. The result is that high-cost development is subsidized by low-cost development with the perverse effect that over time there is more high-cost and less low-cost development (see also Blais 2010). Local community costs increase, as must local taxes and other revenues to pay for them. The long-term outcome is inefficient development patterns.

Table 1.2 illustrates this for Albuquerque, New Mexico. During the middle 2000s, studies in Albuquerque showed that little or no new investment was needed to accommodate growth in some parts of the city, while other parts required new facilities to accommodate new development (see Nelson, Bowles, et al. 2008). City-operated facilities include public safety, parks, recreation and trails, drainage, and streets. Table 1.2 shows the difference in "net" costs needed to serve the same kind of development in what the city calls its fully served area but which is better characterized as the infill/redevelopment area of the city compared to what the city calls its "partially served" area, which is better characterized as the greenfield area of the city. Net costs are the new costs to service new development less the new tax and fee revenue it generates. Clearly, it would cost the city much less to accommodate development in the infill/redevelopment area than in the greenfield area. Unfortunately, its facility financing system does just the opposite by subsidizing higher-cost development by overcharging lower-cost development. Albuquerque thus gets more high-cost and less low-cost development as a result.

MAXIMIZING LAND-USE INTERACTIONS

I also showed in Reshaping Metropolitan America (Nelson 2013b) that more densely developed areas with more mixed uses and more transportation options were more productive than less densely developed areas with segregated land uses and few alternatives to the automobile. A key purpose of redevelopment is to increase development density and to broaden the land-use mix, ideally taking advantage of existing, new, or planned transit systems.