Rather, each city has a place and a role as an element in a system of cities, which serves a functioning economy and geographic region.

The nature of the system of cities has implications for the nature of the space markets underlying the real estate assets in those cities.

To take a simple example, if you try to place in a regional center, such as Des Moines, Iowa, a high-rise upscale apartment building, such as would work in a large financial center, e.g., New York, you probably will not succeed.

The type of real estate that works in the space market depends on the size and economic characteristics of the city in which it is located.

Perhaps the most basic physical characteristics describing cities are size (in terms of population) and location.

Of these two, city size is perhaps the simplest characteristic to describe and measure.

Yet the population of a city, and how that population changes over time, has important impacts on its space markets.

Larger, more rapidly growing cities will generally be able to sustain higher real rents, other things being equal.

With this in mind, let us begin our examination of cities with a look at the patterns that exist in the sizes of cities.

In fact, the sizes of cities tend to follow a rather striking pattern.

This pattern is known as the rank/size rule (also referred to as Zipf's Law).

The rank/size rule can be expressed as follows:

City population = Largest city's population / Rank of city

In other words, if you rank all the cities in a system of cities (that is, all the cities in a coherent geographical region), such that the most populous city has a rank of 1, then the population of each city will approximately equal the largest city's population divided by the rank of the city.

Now let's study the city location.

According to the distribution map of the United States, the three largest cities are far apart, each in a different region of the country: New York in the East, Chicago in the Midwest, and Los Angeles in the West.

With some exceptions, cities of similar size are not located near each other geographically.

Similarly, China and India, you will also note the same pattern.

The rank/size rule and the geographical regional dispersion noted earlier together suggest a hierarchical structure of cities and a division of the U.S. territory into zones of influence.

This, in turn, suggests two points:

Centralizing city-causation (centripetal) forces are counterbalanced by opposing decentralizing (centrifugal) forces.

The relative strength of the centralizing and decentralizing forces differs for different functions and activities.

The centralizing forces cause cities to form in the first place and to tend to agglomerate into fewer and larger cities over time.

The decentralizing forces tend to limit the growth of city size and tend to make more, smaller cities dispersed throughout the territory.

Three primary centralizing forces cause cities to coalesce: economies of scale, economies of agglomeration, and positive locational externalities.

Operating together, these three forces underlie an important characteristic of urban growth dynamics known as cumulative causation.

What is meant by these fundamental principles of urban dynamics?

Economies of scale refers to the phenomenon whereby it is cheaper and more efficient to produce more of a good or service in larger volume at fewer sites.

Fundamentally, economies of scale are caused by fixed production costs at a given site.

The fixed costs are independent of the quantity produced at that site.

The physical plant and equipment in an automobile factory are examples.

As a result of scale economies it is typically more efficient to produce 200,000 automobiles per year in one factory than to produce 50,000 automobiles per year in each of four factories at different locations.

Economies of agglomeration are cost or productivity advantages to the clustering of firms or work sites physically near each other.

Unlike scale economies, which require increasing quantity of production to obtain greater efficiency, agglomeration economies do not require increased quantity of production per se, but merely the clustering of production work sites.

Agglomeration economies result from the ability of different types of workers and production processes to help each other if they are physically near each other.

This may result from vertical linkages (upstream and downstream linkages in production processes) or from horizontal linkages (synergy across firms or work sites, such as the sharing of a common pool of skilled workers or the existence of a critical mass of experts in a given field, resulting in a creative cross-fertilization and stimulation of individuals).

A closely related economic concept is that of positive locational externalities.

This occurs when one firm benefits from the nearby location of another firm without the first firm being able to capture all of these benefits for itself.

This concept is similar to agglomeration economies.

The difference is that locational externalities may occur between as few as two firms or plants, not necessarily requiring a critical mass of firms or workers.

An example of a positive locational externality would be the benefit obtained by a trucking firm whose operations are centered, say, in Nashville, Tennessee, when a major airline establishes an airfreight hub in that city.

The airfreight operations will increase demand for the trucking company's services.

A more subtle example of external economies occurs in the interaction of scale economies and agglomeration economies and the interaction of supply and demand.

As more firms cluster together, the local population and production swells, providing a larger local market for all manner of goods and services.

This allows larger-scale production, which, in the presence of scale economies, allows lower cost and more efficient production.

This, in turn, stimulates further demand for additional production.

Thus, in the presence of scale economies, firms can experience external benefits merely from their proximity to each other, even if they are not directly technologically and physically related to one another.

In other words, scale economies can integrate the notions of agglomeration economies and positive locational externalities, at least for goods and services that are locally produced and distributed.

The sort of growth spiral just described has been referred to as cumulative causation.

It suggests that there can be momentum in the growth (or decline) of cities and regions.

Growth breeds growth (and decline breeds decline).

If we look at the overall system of cities in the United States and define growth and decline, relatively speaking, by the rank of a city's population, we see that when a city commences growing, it tends to continue growing for at least a few decades; and vice versa: when a city begins to decline, the decline tends to persist for at least a few decades.

Centrifugal forces are factors that cause population and economic activity to disperse.

The decentralizing forces that put a brake on urban agglomeration and result in a larger number of smaller cities include factors that make it less efficient or more costly to produce goods and services in large cities, or that reduce the quality of life of the inhabitants.

For example, congestion, pollution, crime, high intra-urban transportation costs, as well as high rents and urban-land costs, are some of the negative features of large cities that tend to get worse as cities increase in size (at least beyond some critical limit).

Also, a larger average size of cities relative to the total population of a country implies that the number of cities in the population must be fewer.

For example, a country with 100 million city dwellers can have at most only 10 cities with 10 million inhabitants, but it could have as many as 100 cities with 1 million inhabitants.

Spread out over the same geographical territory, the 10 "supercities" would be on average much farther apart from each other and from the average rural point in the territory than would 100 cities of 1 million inhabitants each similarly spaced throughout the same territory.

Thus, larger (and therefore fewer) cities will tend to increase rural-to-urban transport costs, due to greater average transport distances (possibly offset by reduced city-to-city transport requirements).

When centralizing forces become more powerful or more important in a society, one or a few large cities tend to siphon population and grow faster than smaller cities.

For example, reductions in transport and communication costs and the rise of highly centralized governmental and administrative institutions resulted in the development of dominant capital cities in imperial nations, such as London, Paris, and Moscow during the 19th century and the first half of the 20th.

These cities often had larger populations than the rank/size rule would predict.

When decentralizing forces become more powerful or important, small- or medium-size cities will capture most of the population growth in a society.

For example, in the United States in the late 20th century, medium-size cities generally grew faster than the largest cities.

A fundamental key point about the role of centralizing and decentralizing forces in shaping the system of cities is that centralizing forces are relatively stronger than decentralizing forces for some types of activities.

For example, upper-level national governmental functions and international financial services are highly concentrated in only a few of the largest and most expensive cities.

For these functions, the centralizing forces must be relatively strong, or they would not concentrate in such high-cost locations.

In contrast, light manufacturing, distribution, corporate and governmental research, and lower-level governmental and corporate-managerial functions are widely spread out among cities, including many small- to medium-size cities.

For these activities, the centralizing forces must be relatively weak, and/or the decentralizing forces relatively strong.