As we near the end of the course, we turn our interest from the actions of individual landowners, farmers, and industrialists to cities and regions as economic units.  This particular lesson reviews a common approach to understanding the economic growth of a region (such as a metropolitan area, a state or province, or even a small country) as a function of the growth in exports from the region.

PROJECTING  REGIONAL  GROWTH

It's often useful to be able to project the population and economy of a region, especially for public- and private-sector infrastructure investment, which takes time to plan, fund, design, and build.

• School districts need to plan their building needs for 5-10 years in advance.
• Transportation planners need to understand the likely demands on the transportation system at least 10 years in advance.
• Local retailers want to plan their expansion (or contraction), and non-local retailers want to identify regional markets to enter.
• Real-estate developers and their financiers need estimates of the demand for new construction.

How can we make these projections?

THE  ECONOMIC  BASE  OF  A  REGION

One commonly-used approach to forecasting regional economic growth and labor employment is to forecast the external demand for goods and services produced in the region, and then to multiply changes in that external demand by a factor to come up with the total demand for goods and services -- for export and for local consumption.
This assumes that the region's economy is driven by external demand. 

• This makes some sense for small areas, like metropolitan areas or states.
• It makes less sense for huge national economies like the U.S.  Recall that exports represented just $1 trillion or 12% of the U.S.'s 8.1 trillion-dollar economy in 1997.
• Think about the extremes:  none of the income of the world as a whole comes from extra-terrestrial exports, while all the income of your household probably comes from extra-household exports (primarily of labor).
• You can't forecast the size of the global economy in 2002 based on the likely size of export income to the world, but you can probably forecast the size of your household economy in 2002 based on the likely size of the labor income earned outside the household by people in the household.

To accomplish this, arbitrarily divide the total quantity of regional production (Q) into export-oriented and locally-oriented components.  We call the export-oriented component the basic output (Q_b) and the locally-oriented component the non-basic output (Q_n):

Sometimes we can forecast changes in Q_b  more easily than changes in Q or  Q_n:  often, there are relatively few large externally-oriented producers, and we can either survey them directly or study trends in national demand for those kinds of goods and services.  (What would the key externally-oriented producers be for metropolitan Seattle?  for Washington State overall?)

If we can forecast changes in Q_b , how would we apply them to get a forecast of the change in the total economy?  If we assume that the ratio between Q_n and Q_b  remains constant, it's pretty easy.

Another way of writing this is Q = Q_b ( 1 + a ) , which is what you need to remember.

What is  Q = Q_b ( 1 + a )  telling us?  It says that the region's total production is equal to its basic (exported) production times a multiplier:  that multiplier is  1 + a, which is the same as 1 + Q_n / Q_b .
 

If we know that forecast change in  Q_b   -- let's call it  cQ_b   -- then cQ = cQ_b ( 1 + a ) .

We'd add that to the current Q, so that the future total production, Q^t+1 , equals  Q + cQ  , or

Often, you'll read newspaper accounts such as:
 

Let's take a production figure like $150 million per year (that's not an unreasonable figure for a manufacturing operation that employs 2,000 workers).  This is basic economic activity -- most of the demand for these bubble widgets is outside the county and even outside the state.  The analysts might have a multiplier from some previous study of manufacturing output and total output in the state -- let's say the multiplier (1 + a) is 2.5.  That's saying that every $1 million in basic production calls forth another $1.5 million in nonbasic production, for a total of $2.5 million in total production.  So the total change in the state's production, with the new plant, is cQ = cQ_b ( 1 + a ) = $150M  (2.5) = $375M per year.

Are the state's promised expenditures worthwhile?  Well, how would the state be "repaid"?  Not from corporate taxes paid by MegaWidget, at least not for the first 10 years.  However, each of the new jobs represents income tax to the state (if the state has personal income tax) and sales taxes to the state (since most of those workers are going to buy most of their household goods and services in the state).  How much?  We've got to make some assumptions, based on current ratios that the state analysts would know.

I'm going to translate the $375M increased state production into 6,500 additional jobs (because some of the nonbasic activities produce more jobs per dollar of output than does capital-intensive widget manufacturing -- that's why some of these non-basic activities don't pay as much).  If you average the wages paid by facilities like MegaWidget and by the other, non-basic, activities, you might get $25,000 per job per year.  That's $162.5 million of additional, annual, personal income in the state.  If state income and sales taxes capture 8 percent of that additional personal income, that's $13 million in extra state revenues per year.  Wow!

The governor has agreed to spend $115 million up front.  Is this worthwhile?  Well, 30 years of $13 million per year is $390 million in additional state revenues (30 years is the maximum expected life of MegaWidget's investment).  However, if the state has to float bonds to pay for the $115 million of expenditures, there's a hefty interest cost.  It's not unlikely that the principal and interest payments on the bonds would equal the $390 million.  Then, there are the additional costs of educating the children of those additional workers....

Let's assume the deal goes through.  We're talking 6,500 additional jobs in the county.  How many people is that -- what does this do to net migration?

That depends on how many in-migrants come in to take those jobs, and how large are their households.  We'd have to have some more information to make a good projection, but I'll hazard a guess that 4,000 of those jobs will go to in-migrants, representing 3,000 new in-migrant households (some households have more than one worker) and, at 2.5 people per household, 7,500 new people in the county in a few years.

If Hungry County has 100,000 people now, that's a 7.5 percent increase -- large enough to be noticeable by real estate developers (3,000 additional housing units), the school district (a few thousand additional children), and transport planners (6,500 additional people traveling to work, 3,000 additional households driving kids to events and trying to run errands on Saturday).
 

DETERMINING  THE  ECONOMIC  BASE  OF  A  REGION

How do we know how much of the regional economy is basic, and how much is non-basic?

• We can survey the producers in the region, asking how much of their production is sold to firms, governments, and households in the region versus outside.
• We can assume that most of the production and employment in some sectors is basic (exported), and most of the production and employment in other sectors is non-basic (purchased internally).  Often, we assume that most large-scale agriculture, mining, manufacturing, and certain services (packaged software, for example) are basic;  and that most construction, retail/wholesale services, and state-local government are non-basic.
• We can identify the sectors that produce a lot more than the region is likely to use internally, and call their output basic (export oriented).

Location Quotients
A location quotient is a way of expressing how specialized a region is in a certain sector.  This is accomplished by relating the proportion that a sector represents of the region's total production (or employment) to the proportion that the sector represents of the nation's total production.   Arithmetically,

We generally assume that any sector with a location quotient substantially above 1.0 is at least partially a basic sector in that region's economy.

If we use total output by sector for Washington State and the U.S. as a whole in 1994 (U.S. Bureau of Economic Analysis data), we get the following location quotients for Washington State sectors.  Which sectors seem to be part of Washington's economic base?
 

All this attention to the economic base is useful if we can, in fact, forecast the economic base of a region.  Once we've identified which sectors are basic, how might we forecast their output (or employment)?

• We could rely on announcements of expansions or contractions in key sectors.  This will give us spotty coverage, however.
• We could survey business and government leaders in those sectors, to learn their projections for growth or decline.
• We could use national projections for output in each sector that is part of our region's economic base.  We could then assume that the region will maintain its current share of forecast national output by sector.
• We could study the trends in the region's share of national output by sector, and use those trends to forecast the future.

Hoover and Giarratani (An Introduction to Regional Economics, third edition, Chapter 11.3.1) provide caveats about this concept of regional growth:
 1) For one thing, there is no reason to expect the multiplier to remain constant over any long study period.
 2) "We are left with the implication that a region will grow faster if it can manage to import less, and that growth promotion efforts should be directed toward creating a 'favorable balance of trade'"  Note from international accounts that a trade (or current-account) deficit implies a net capital inflow, which is not necessarily a bad thing.  I.e., a negative trade balance for a given region means that the region is being lent resources from other regions:  if these resources are being invested for future productivity increases, that's a good thing for the region.  If the future increase in regional productivity is higher than that net investment would result in within other regions (e.g., the region that is lending the resources by carrying a trade surplus with our region), then the entire system benefits.