Leontief Input-Output Economic Model

Objectives 1. To investigate the Leontief Input-Output Economic Model. 2. To use MATLAB to solve open and closed Leontief economic systems. Prerequisites 1. Systems of linear equations and their solutions in MATLAB. 2. Inverses of matrices. 3. Eigenvectors and eigenvalues (for exercise 4 in the Laboratory Section). Background In 1949, Wassily W. Leontief developed an economic model of the United States economy by dividing it into 500 economic sectors. He developed a linear equation for each sector representing how that sector distributes its outputs to the other sectors of the economy. In the Leontief input-output model, the economic system is assumed to have n industries with two types of demands on each industry: external demand (from outside the system) and internal demand (demand placed on one industry by another in the same system). We assume that there is no over-production so that the sum of the internal demands plus the external demand equals the total demand for each industry. Let xi denote the i'th industry's production, ei the external demand on the ith industry, and aij the internal demand placed on the ith industry by the jth industry. This means that aij is the number of units of the output of industry i required to produce 1 unit of industry j's output. The total amount industry j needs from industry i is aijxj. Under our assumption that the total demand is equal to the output of each industry, we have the following linear system: a11 x1 + a12 x2 + + a1n xn + e1 = x1 a21 x1 + a22 x2 + + a2 n xn + e2 = x2 an1 x1 + an2 x2 + + ann xn + en = xn Rewriting in standard form, we have: (1 a11 ) x1 a12 x2 a1n xn = e1 a21 x1 + (1 a22 ) x2 a2n xn = e2 an1 x1 an2 x2 + (1 ann ) xn = en The matrix A with entries aij is often called the technology matrix. Note that the matrix form of this system is (In-A)x = e. The matrix In -A is called the Leontief matrix. The system can be solved in MATLAB using the same techniques that we have been using. For example, consider an economic system with three industries each having external demands 10, 24, and 20 respectively. Also assume that a11 = 0.2, a12 = 0.5, a13 = 0.15, a21 = 0.4, a22 = 0.1, a23 = 0.3, a31= 0.25, a32 = 0.5; and

a33 = 0.15.. To find the outputs x1, x2, and x3 of the three industries we set up the system in standard form as indicated above (using 1- a11 = 0.8; 1-a22 = 0.9, and 1a33 = 0.85): 0. 8 x1 0.5 x2 0.15 x3 = 10 0. 4 x1 + 0. 9 x2 0. 3 x3 = 24 0. 25 x1 0. 5 x2 + 0. 85 x3 = 20 The following MATLAB session finds the solution:
L = [0.8 -0.5 -0.15; -0.4 0.9 -0.3; -0.25 -0.5 0.85] ; b = [10; 24; 20] AUG = [L b] AUG = 0.8000 -0.5000 -0.1500 10.0000 -0.4000 0.9000 -0.3000 24.0000 -0.2500 -0.5000 0.8500 20.0000 rref(AUG) Since the solution is unique, we can use either ans = rref (AUG) or L\b. 1.0000 0 0 108.0408 0 1.0000 0 115.8324 0 0 1.0000 123.4429 L\b ans = 108.0408 115.8324 123.4428

This tells us that approximately 108 units from industry 1, 116 units from industry 2, and 123 from industry 3 are needed for supply to equal demand. In the laboratory exercises that follow you will investigate several different problems for economies using the Leontief input-output model and different techniques for finding solutions.

Laboratory on Leontief Input-Output Economic Model

c) Compute the output levels required of each industry to meet the internal and external demands. Industry 1 __________ Industry 2 _____________ Industry 3 __________ 2. Another way of computing the solution to the system (I n -A)x = e is to explicitly compute (I nA)-1 and then calculate x = (In-A)-1e. Use this method for the following problem. Assume there are five industries with the following technology matrix: 1 2 3 4 5

Steel 1 Petroleum 2 Electricity 3 Coal 4 Agriculture 5

0.20 0.10 0.15 0.10 0.10

0.30 0.10 0.15 0.15 0.20

0.10 0.10 0.05 0.05 0.05

0.20 0.20 0.40 0.05 0.20

0.10 0.05 0.01 0.0 0.20

Fill in the following table with the calculated output levels needed for each of the five industries to meet the external demands for each (in millions of dollars) given in the table. Steel External Demand Output Level External Demand Output Level External Demand Output Level 30 Petroleum 30 Electricity 40 Coal 20 Agriculture 50

10

15

30

25

40

15

20

45

25

35

3. Efficient calculations. For a large technology matrix A that is needed in realistic Leontief economic models, calculating the inverse of I n -A involves a lot of computation. Since A will generally have very small nonnegative entries, for large enough m,

(I A) 1 I + A + A 2 + + A m

This is justified as follows:

Since A's entries are small, Am approaches the zero matrix as m gets larger. Hence

( I A)(I + A + A 2 + + A m ) =

I ( I + A + A 2 + + A m ) A ( I + A + A 2 + + A m ) = (I + A + A 2 + + A m ) ( A + A 2 + A 3 + + A m+1 ) = I A m +1 I
a) For the matrix in exercise 2, verify that A m is approaching the zero matrix by calculating A^10, A^20, A^30. b) Calculate I1 = I + A + A 2 + + A m for m = 10, and compare your answer to the answer computed by MATLAB's inv(eye( n)-A). Use the following commands: I1 = eye(5)+A +A^2+A^3+A^4+A^5+A^6+A^7+A^8+A^9+A^10 I2 = inv(eye(5)-A) max(max(I1-I2)) (The last command finds the maximum of the row, column entries.) Largest difference = ___________ c) The calculation in b) is inefficient because it repeats calculations of lower powers of A to calculate higher powers of A. The M-file function horners(A,m) calculates the same more efficiently, using an algorithm called Horner's method. Examine its code (Choose File-Open --C:\MATLAB\karend\horners.m), and explain why it is more efficient.

d) Verify its efficiency by comparing the number of floating point operations for the straightforward method and for Horner's method, using the commands: flops(0) eye(5)+A+A^2+A^3+A^4+A^5+A^6+A^7+A^8+A^9+A^10 flops flops(0) horners(A,10) flops Flops (straight-forward method) ________ Flops (Horner's method) _________

3. Leontief Economic Model for the U.S. 1958 economy . 1 The following table gives the technology matrix for the 1958 U.S. economy based upon Leontief's figures: Final Nonmetal Final Nonmetal Final Metal Basic Metal Basic Nonmetal Energy Services 0.170 0.003 0.025 0.348 0.007 0.120 Final Metal 0.004 0.295 0.173 0.037 0.001 0.074 Basic Metal 0 0.018 0.460 0.021 0.039 0.104 Basic Nonmetal 0.029 0.002 0.007 0.403 0.025 0.123 Energy 0 0.004 0.011 0.011 0.358 0.173 Services 0.008 0.016 0.007 0.048 0.025 0.234

The external demands of the economy are listed as ( in millions of dollars): Final Nonmetal 99,640 Final Metal 75,548 Basic Metal 14,444 Basic Nonmetal 33,501 Energy 23,527 Services 263,985

Enter the technology matrix A, the external demand vector e, and calculate the inverse matrix (IA)-1, and the output vector x = (I-A)-1 e. Write down the output needed for each sector (in millions of dollars) in the table that follows: Final Nonmetal Final Metal Basic Metal Basic Nonmetal Energy Services

b) Calculate the change in the output under the following assumptions: i) External demand for Energy decreases by 10%. Final Nonmetal Final Metal Basic Metal Basic Nonmetal Energy Services

ii) External demand for Services increases by 10%. Final Final Metal Basic Metal Basic Nonmetal Nonmetal

Energy

Services

1From

Grossman, Stanley I., Elementary Linear Algebra , 5th Edition.

4. Closed Leontief economic model . In the closed Leontief input-output model, it is assumed that there is no demand from an open sector (that is, the vector e is the zero vector). The problem is then to find outputs to satisfy the demands of the n industries. The resulting system of equations can be represented as (A-I) x = 0. With no open sector, the total inputs for industry j should be the same as the output for industry j, so that a1 j x j + a2 j x j + + anj x j = x j , and a1 j + a2 j + + anj = 1. The solution x to (AI)x = 0 will be the eigenvector associated with eigenvalue 1. (It can be shown using the PerronFroebenius Theorem that the matrix A will always have 1 as an eigenvalue with a corresponding eigenvector that has all positive entries.) 0.20 0.35 0.10 0.20 Assume the closed model for four industries with technology matrix A given on the left. Use [ V D] = 0.15 0.12 0.32 0.10 eig(A) to calculate the eigenvectors for A and verify A = 0.30 0.40 0.25 0.45 that 1 is an eigenvalue. Find the corresponding 0.35 0.13 0.33 0.25 eigenvector.

The solutions to (A-I)x = 0 will be (positive) scalar multiples of this eigenvector. a) Write down the form for the general solution.

b) What must be produced by each of the other industries if the output for the first industry, x1, is equal to 1000? x2 ___________ x3 __________ x4 __________