Lecture 4 Firms and Production

Firms and Production Lecture 4 August 2015 1 / 62 Introduction In this lecture we look at …rms and production. It is the …rst step in deriving the supply curve we say in the …rst lecture. 2 / 62 Outline Ownership and Management of Firms - What exactly is a "…rm?" Production - How a …rm makes output from their set of inputs. Short-Run Production - Look at production when the …rm has a …xed input. Long-Run Production - Look at production when there are no …xed inputs. Returns to Scale - How the size of a …rm a¤ects how much it produces. Productivity and Technical Change - The most output you can get for your inputs varies across …rms and across time. 3 / 62 Ownership and Management of Firms A …rm is simply some organization that takes inputs and turns it into outputs. We can roughly divide these …rms into private sector public non-pro…t …rms 4 / 62 Ownership and Management of Firms Sole proprietorship owned by an individual who is responsible for all debts. Example: A freelance writer or a bookkeeper. 5 / 62 Ownership and Management of Firms General partnership Jointly owned by multiple people who are together responsible for debts. Example: Law o¢ ce with multiple partners. 6 / 62 Ownership and Management of Firms Corporations Owned by shareholders in proportion to the amount of stock they own. limited liability. Example: Microsoft, recently Facebook. 7 / 62 Ownership and Management of Firms Private …rms have the single goal of maximizing pro…ts. Pro…t (π ) is de…ned as total revenue (TR ) minus total cost (TC ). π = TR TR TC = p q A …rm can only maximize pro…t if it achieves technical e¢ ciency. Technical e¢ ciency means they get the most output they possibly can from their set of inputs and technology. 8 / 62 Production A …rm takes inputs (factors of production) and turns it into output according to its technology. We can broadly classify these inputs into capital (K), labour (L) and materials (M) (which we usually ignore for simplicity). 9 / 62 Production The production function summarizes this process, and tells us exactly how much output the …rm can get from their inputs. For example suppose our production function is q = f (L, K ) = 2 L K If the …rm employs two units labour and 4 units of capital it gets 16 units of output (it could produce less, but that would not be e¢ cient). 10 / 62 Production EXAMPLE Suppose a …rm produces output using only labour according to the production function q = L2 . Sketch this production function and identify two production plans, one that is technically e¢ cient and one that is not. 11 / 62 Production The short run is the period of time that at least one factor of production cannot be changed. For example, dominoes can decide how many delivery drivers it hires in a month, but can’t decide how many stores to build in this time frame. 12 / 62 Production The long run is a period of time in which all inputs can be varied. The di¤erence between short and long run varies by industry. We call inputs that can’t be changed …xed inputs, and ones that can be changed variable inputs. 13 / 62 Short-Run Production Lets say capital is …xed in the short run, our production function is then q = f (K , L) Suppose our production function is q = 2KL, but capital is …xed at K = 4 in the short-run. Our short-run production function becomes q = 8L. We can summarize the relationship between output and the amount of labour used by the total product of labour, the average product of labour and the marginal product of labour. 14 / 62 Short-Run Production Total product of labour is the amount of output that a given amount of labour can produce holding other inputs …xed Marginal product of labour is the extra output you get from increasing labour by some in…nitesimally small amount MPL = ∂q ∂L Average product of labour is the amount of output produced per worker q APL = L 15 / 62 Short-Run Production EXAMPLE Suppose our production function is Q=K L2 What is the total product of labour, marginal product of labour and average product of labour if capital is …xed at 50? 16 / 62 Short-Run Production If our short run production function is q = L + 30L2 L3 17 / 62 Short-Run Production We assume the …rm can hire fractions of workers (which is why this is smooth) The MPL is the slope of the production function The APL is the slope of the chord from the origin 18 / 62 Short-Run Production The previous production function is typical APL initially rises because of gains from specialization, but it declines because capital is held constant. 19 / 62 Short-Run Production The MPL always intersects the APL at the maximum of the APL curve. When MPL > APL , the average is pulled up. When MPL < APL , the average is pulled down. 20 / 62 Short-Run Production EXAMPLE Let’s prove together that MPL = APL at the maximum of APL for the production function q = L + 30L2 L3 . 21 / 62 Short-Run Production The law of diminishing marginal returns is huge in economics. As you increase one input, holding all other inputs and technology constant, the marginal returns to that input will decrease eventually. The second derivative will become negative. You can’t grow the world’s food supply in a ‡ower pot. This is why Malthus predicted mass starvation (one input -land- is …xed). 22 / 62 Short-Run Production For the production function q = L + 30L2 L3 dq = 1 + 60L MPL = dL 3L2 23 / 62 Short-Run Production The marginal product of labour is increasing for low levels of output because of gains from specialization. Eventually, the marginal product of labour starts to decrease. dMPL = 60 6L dL L < 10 MPL increases L > 10 MPL decreases 24 / 62 Short-Run Production EXAMPLE Suppose the world’s supply of food is determined by the production function 1 1 F = M 2 L2 M is land and is …xed at 3600. L is labour. At what point does this production function exhibit diminishing marginal returns to labour? 25 / 62 Long-Run Production In the long run, the …rm is free to select as much of any input... nothing is …xed. Lets suppose our production function is Cobb-Douglas 1 1 q = 3L 2 K 2 Try to graph this. 26 / 62 Long-Run Production Now that we have multiple variable inputs, our production function has multiple dimensions. Can’t really deal with it graphically. We can summarize this production function in 2 dimensions. 27 / 62 Long-Run Production An isoquant is a curve that shows the e¢ cient combinations of inputs that produce a single level of output This has a very similar interpretation to indi¤erence curves. Every combination of labour and capital on the same isoquant will produce the same amount of output. 28 / 62 Long-Run Production 29 / 62 Long-Run Production Say our production function is Q = 10 K L Draw the isoquant for Q = 10 and Q = 20. 30 / 62 Long-Run Production Lets now discuss the properties of isoquants. 1. Further from the origin, the greater the output The more inputs you use, the more output you get if you are producing e¢ ciently 31 / 62 Long-Run Production 2. They cannot cross Suppose the isoquant where Q = 20 and Q = 15 cross. The …rm could produce 15 or 20 units of output for the same input combination. The …rm would not be e¢ cient if it produced 15 for that input combination. 32 / 62 Long-Run Production 3. They slope downward If they sloped upward, the …rm could produce the same level of output with fewer inputs. 33 / 62 Long-Run Production 4. They are thin If they were thick, the …rm could decrease its input use and get the same level of output 34 / 62 Long-Run Production 5. UNLIKE indi¤erence curves, isoquants are a cardinal measure. Output is objective, it is not some abstract thing like utility. 35 / 62 Long-Run Production The curvature of the isoquant tells us how substitutable/complementary the inputs are in the production process. The more "curvy" the isoquant, the greater degree of complementarity there is between inputs 36 / 62 Long-Run Production If they are straight lines, the inputs are perfectly susbstitutable (apples from Oregon or Washington). This would come from a linear production function q = x +y If they are right angles, inputs must be used in …xed proportions (one secretary per phone). this comes from a …xed-proportions production function q = minfx, y g 37 / 62 Long-Run Production 38 / 62 Long-Run Production The slope of the isoquant shows the …rm’s ability to replace one input with another holding output constant. This is called the marginal rate of technical substitution (MRTS ) . How much K can we give up for another unit of L holding output constant. 39 / 62 Long-Run Production Remember that the slope of an indi¤erence curve is the negative ratio of marginal utilities. The slope of the isoquant (MRTS ) is the ratio of marginal products. MRTS = dq dL dq dK = MPL MPK 40 / 62 Long-Run Production EXAMPLE Lets …nd the MRTS of q = ALα K 1 α Is the MRTS constant for this production function? 41 / 62 Long-Run Production If we have normal convex isoquants, we have a diminishing marginal rate of substitution That is, when a lot of capital and little labour is used, the MRTS is really high. When a lot of labour and a little capital is used, it is really low. 42 / 62 Long-Run Production Suppose we have a factory with 1,000,000 workers and only one machine. To keep output constant, we could trade one machine for a ton of workers (because workers need machines) As we get more machines, the number of workers we can trade for one more machine will decrease. 43 / 62 Long-Run Production 44 / 62 Long-Run Production For most isoquants, the MRTS is not constant. As we increase capital and decrease labour, at what rate does the MRTS change? This is the elasticity of substitution... A measure of how "curvy" our isoquants are 45 / 62 Long-Run Production Elasticity of substitution (σ ) is the percentage change in the capital labour ratio w.r.t a percentage change in the MRTS σ= d (K /L ) K /L dMRTS MRTS = d (K /L) MRTS dMRTS K /L If σ is really high, that means a tiny change in the MRTS results in a big change in K /L, the isoquant is pretty ‡at. 46 / 62 Long-Run Production EXAMPLE What is the elasticity of substitution for the following Cobb-Douglas production function? Q = ALα K β What is the elasticity of substitution for a linear production function? What does that mean? 47 / 62 Returns to Scale If we increase our inputs proportionately, what happens to our output? This is called returns to scale We can have increasing, decreasing or constant returns to scale. 48 / 62 Returns to Scale If doubling our inputs leads to exactly double the output, we have constant returns to scale 2f (L, K ) = f (2L, 2K ) 49 / 62 Returns to Scale If doubling our inputs leads to more than double the output, we have increasing returns to scale 2f (L, K ) < f (2L, 2K ) Could be caused by greater specialization 50 / 62 Returns to Scale If doubling our inputs leads to less than double the output, we have decreasing returns to scale 2f (L, K ) > f (2L, 2K ) Could be caused by management or organizational problems 51 / 62 Returns to Scale What industries do you think have constant, increasing or decreasing returns to scale? What are some factors that determine returns to scale? 52 / 62 Returns to Scale Generally speaking, our production function is homogenous of degree γ when f (xL, xK ) = x γ f (L, k ) 53 / 62 Returns to Scale EXAMPLE The following production function is homogenous of degree what? 1 1 Q = K 2 L4 54 / 62 Returns to Scale It is possible for production functions to have varying returns to scale. Could have increasing returns to scale for low levels of production, and decreasing returns to scale for high levels of production. For low levels of production, you have gains to specialization and for large levels of production, you run into management problems. 55 / 62 Returns to Scale 56 / 62 Productivity and Technical Change Even if two …rms are producing e¢ ciency, it is possible that they are not equally as productive. We can express a …rm’s relative productivity by the ratio of the …rm’s output q to the amount of output the most productive …rm in the industry could have produced from the same inputs q ρ= q q 100 If you are the most productive …rm in the industry, what is ρ? Estimated that the average productivity of manufacturing …rms in the US is 63% to 99%. 57 / 62 Productivity and Technical Change It is possible that one …rm can produce more today from a given amount of inputs than it could in the past. An advance in knowledge that allows more output to be produced from the same level of inputs is called technical progress. Can be neutral or non-neutral. 58 / 62 Productivity and Technical Change Neutral technical change means the …rm can produce more output using the same ratio of inputs. q = A(t )f (L, K ) For example q1 = 10 K .5 L.5 q2 = 100 K .5 L.5 59 / 62 Productivity and Technical Change Or it can be non-neutral in which innovations alter the proportion of input used. q1 = 10 K .5 L.5 q2 = 10 K .5 L.8 If a machine is invented that requires only one person to operate it rather than two, this is a non-neutral labour saving technical change. 60 / 62 Summary What is technical e¢ ciency? What does a production function show you? What is the di¤erence between the short and long run? 61 / 62 Summary What causes diminishing marginal returns? What is an isoquant What is the marginal rate of substitution and the elasticity of substitution? What are returns to scale? What is the di¤erence between a neutral and non-neutral technical change? 62 / 62