Chapter 11: Behind the Supply Curve: Inputs and Costs

Welcome to Last Minute Lecture.

This free chapter overview is designed to help students review and understand key concepts.

These summaries supplement not replaced the original textbook and may not be redistributed or resold.

For complete coverage, always consult the official text.

Welcome to the Deep Dive, your shortcut to being well informed.

Today we're pulling back the right.

And, you know, to kick us off, let's think about that classic image from America, the beautiful,

those amber waves of grain.

A powerful symbol.

It really is.

But here's a question for you, maybe a surprising one.

Why might European farmers often get,

say, three times as much wheat per acre as US farmers, even when their skills are pretty similar?

Yeah, that's a great question.

It's fascinating, isn't it?

And the answer isn't really just about better farming techniques.

It really boils down to economic decisions, specifically decisions made at the margin.

Meaning how firms decide to use their inputs based on the incentives they face and what that implies for their costs.

Think about European policies like price floors for wheat.

Okay, guaranteed prices.

Exactly.

That gives farmers a really strong incentive to, well, throw more variable inputs at the problem, more fertilizer, more labor, really intensive work per acre.

So they spend more.

They do spend more.

Their costs go up.

But those guaranteed higher prices make that extra spending worth it for them.

It's a really clear example of how something external like policy shapes a firm's production choices, their costs, everything.

Which is exactly what we're digging into today.

Precisely.

So our mission in this deep dive is simple.

Give you a clear, comprehensive kind of conversational summary of these core ideas, how firms produce things and what it costs them.

We're basically distilling a key microeconomics chapter, the kind you'd find in Krugman and Wells.

Turning textbook concepts into practical insights.

Exactly.

Helping you visualize the models, apply the ideas, whether you're prepping for an exam or just curious about what makes the economy tick.

Think of this as your toolkit for getting a grip on firm behavior.

So let's jump in.

Okay.

So to understand firms, we have to start with the basics.

Transforming inputs into output.

You're probably familiar with the idea of a production function.

Right.

Inputs in, output out.

Basically, yes.

It's the technical relationship, the recipe almost, showing the maximum output a firm can get from any given mix of inputs.

And speaking of inputs, there's a key distinction right away, isn't there?

Fixed versus variable.

Absolutely crucial.

Let's use our farm example, Riley and Tyler's Wheat Farm.

Say they have 10 acres of land.

In the short term, they can't just snap their fingers and get more land or less land.

So the land is a fixed input.

It comes with a fixed cost, maybe $400, doesn't matter if they grow a lot or a little wheat.

Okay.

That cost is locked in for now.

Locked in.

But the workers they hire, that's different.

That's a variable input.

They can hire more or fewer.

Exactly.

Day to day, week to week.

And each worker costs something, say $200.

This difference is what defines our time horizons in economics.

The short run and the long run.

Right.

The short run is any period where at least one input is fixed, like Riley and Tyler's land.

The long run is when everything becomes variable.

Given enough time, they could buy more land or sell some off.

All inputs are adjustable in the long run.

Got it.

So how do we visualize this?

You mentioned the production function.

We can graph it as the total product curve.

Imagine a graph.

Horizontal axis is your variable input number of workers.

Vertical axis is your total output bushels of wheat.

Now for those 10 acres, as Riley and Tyler add workers, total output generally goes up, right?

So the curve slopes upward.

More workers, more wheat.

Makes sense.

But here's the key thing.

Look closely, and you'll see that curve starts to, well, flatten out as you add more and more workers.

It's still rising, but slower.

Why does it slow down?

Ah, that flattening, that slowdown.

That shows us our next big concept.

Marginal product.

Marginal.

Meaning the extra bit.

Exactly.

The marginal product of an input is the additional output you get from using one more unit of that input.

So if adding the second worker gives Riley and Tyler an extra 17 bushels of wheat.

That worker's marginal product is 17 bushels.

Precisely.

The marginal product of labor, or MPL.

And technically the MPL is the slope of that total product curve we just talked about.

So a flattening curve means a smaller slope.

Which means a declining marginal product.

Yes.

And this leads us somewhere important, I bet.

It leads us straight to one of the most fundamental ideas in economics.

Diminishing returns to an input.

Okay.

Diminishing returns.

I've heard that phrase.

It means that if you keep adding more of one input, like labor, while holding other inputs fixed, like land, eventually,

the marginal product of that input you're adding will start to fall.

So the first worker adds a lot, the next adds a bit less, the next even less.

Exactly like that.

For Riley and Tyler, maybe worker one adds 19 bushels, worker two adds 17, worker three adds 15, and so are.

Each extra worker contributes less than the one before.

Why though?

Is it just people getting lazy?

Huh?

Not necessarily.

Think about it practically.

The land area is fixed, right?

10 acres.

As you add more workers, each new person has effectively less land to work with.

They start getting in each other's way.

Maybe the most productive tasks are already being done.

The fixed input becomes a bottleneck.

Ah, okay.

That makes sense.

It's about the combination of inputs.

Precisely.

And this is why the marginal product curve, if you plot MPL against the number of workers,

typically slopes downward.

It reflects those diminishing returns.

What if they got more land though?

Say 20 acres.

Good question.

If they increase their fixed input, both their total product curve and their marginal product curve would shift up.

Up.

So they get more wheat overall.

Yes.

Any given number of workers would now produce more wheat because they have more land to work with.

But, and this is important, you'd still see diminishing returns to labor setting in eventually on the larger farm.

The MPL curve would still slope down just starting from a higher point.

Okay.

So diminishing returns is sort of a universal thing when one input is fixed.

Pretty much in the short run.

And this brings us right back to the Europe versus US wheat yield thing.

Right.

The higher European yield.

Those farmers get higher yields per acre, partly because policies incentivize them to use more variable inputs per acre, pushing further along their production function even into that diminishing returns zone because the price makes it pay off.

I see the incentives change the calculation.

Absolutely.

And this idea of incentives and marginal effects, it pops up everywhere.

Think about foreign aid sometimes.

Giving surplus food can seem great, but it can depress local food prices, right?

Which then removes the incentive for local farmers to invest in their own inputs and boost their yields.

It can inadvertently hurt local agriculture.

So cash aid might be better sometimes.

Often, yes, except maybe in immediate famine situations.

Groups like Oxfam often argue for cash because it lets local markets work and keeps those production incentives in place.

Interesting connection.

And you mentioned diminishing returns applies to teams too.

Oh, definitely.

There's research showing that often the ideal team size is maybe four or five people.

Really?

Not bigger.

Beyond that, adding more members often starts to reduce the marginal product of the team.

You get things like social loafing.

It's harder to see who's pulling their weight.

Coordination gets trickier too, I imagine.

Exactly.

More meetings, more emails, more complexities.

Sometimes an extra person can even have a negative marginal product.

They actually slow the team down.

Wow.

So five plus five might not equal ten in teamwork.

Often doesn't.

Two teams of five can easily outperform one team of ten because they avoid those diseconomies, those diminishing returns to collaboration.

Okay, so we've got the physical side down.

Inputs, outputs, diminishing returns.

That's the production function.

What's next?

Money.

Money.

Always comes down to money.

Now we need to translate that physical production into actual dollar costs for the firm.

Right.

So Riley and Tyler's land.

That was a fixed input.

Which means it generates a fixed cost.

FTE.

That $400 for the 10 acres.

It doesn't matter how much wheat they grow.

That cost is fixed.

Often called overhead.

Okay, overhead.

Got it.

And the workers?

They were the variable input.

So hiring them creates variable costs.

VC.

Each worker costs $200, remember?

Right.

So the more workers they hire to produce more wheat, the higher their variable costs will be.

These costs do depend on the quantity of output.

And adding those together gives you?

Total cost.

TC.

Simple enough.

Total cost equals fixed cost plus variable cost.

TC equals SEFC plus VC.

Okay.

Can we graph this too?

Absolutely.

The total cost curve.

This time output quantity is on the horizontal axis and total cost in dollars is on the vertical axis.

And it goes up.

Presumably.

More output costs more.

Always slopes upward, yes.

But notice how it slips upward?

It gets steeper as output increases.

Steeper.

Why steeper?

Think back to diminishing returns.

Because each extra unit of output requires progressively more variable imprint, more worker hours per bushel, the cost of producing each additional unit goes up.

The total cost rises at an accelerating rate.

Ah, it's all connected.

Diminishing returns in production leads to increasing costs.

Well, increasing marginal costs.

Exactly.

You've jumped ahead perfectly.

To make good decisions, firms need to know not just total cost but the cost of the next unit.

That's marginal cost.

MC.

The change in total cost from making one more widget or a bushel.

Precisely.

Calculated as the change in total cost divided by the change in quantity.

And just like MPL was the slope of the total product curve.

MC is the slope of the total cost curve.

You got it.

And since the total cost curve gets steeper, that tells us that marginal cost is generally rising.

Because of diminishing returns again.

Yeah.

Need more inputs for that extra unit so the extra cost is higher?

That's the main reason.

Now, sometimes at very low levels of output, MC can initially fall due to specialization gains.

Workers get better as they do more.

So MC curves often have this kind of swoosh or J shape.

They dip down briefly then rise.

Okay, a little dip then up.

Got it.

What about average cost?

Right, the other key perspective.

Average total cost.

ATC, sometimes it's called average cost.

This is just your total cost divided by the quantity of output.

ATC divided by Q tells you the cost per unit on average.

ATC, TCQ, makes sense.

Yeah.

And what does that curve look like?

This one's famous for its shape.

It's typically U -shaped.

U -shaped?

Why U?

It comes from two opposing forces working against each other.

Let's break ATC down.

Remember TC plus FC plus VC.

So ATC, FC plus VCQ.

We can split that into average fixed cost, AFC plus average variable cost, AVC.

AFC plus AVC equals ATC.

Okay.

Now think about average fixed cost, AFC.

That's fixed cost divided by output, FCQ.

What happens to AFC as output Q gets bigger?

Well, FC stays the same, but Q goes up.

So AFC must go down.

Continuously falls.

We call this the spreading effect.

You're spreading that fixed overhead cost over more and more units.

Okay, AFC always falling.

That pulls the average down.

But then you have average variable cost, AVC.

That's variable cost divided by output, VCQ.

What happens to AVC as output rises?

Because of diminishing returns, variable cost rise faster than output eventually.

Yeah.

So AVC must eventually rise.

Exactly.

The diminishing returns effect causes AVC to rise as output increases.

So we have AFC pulling down and AVC pulling up.

And that's what creates the U -shape for ATC.

At low output levels, the spreading effect, falling AFC is stronger, so ATC falls.

Yeah.

At higher output levels, the diminishing returns effect, rising AVC dominates, and ATC starts to rise.

Making a U and the bottom of that U.

That's the sweet spot, the minimum cost output, where average total cost is as low as it can possibly be.

Okay.

Now for the really critical relationship, you said,

between MC and ATC.

This is absolutely fundamental.

Best way to think about it is like your grade point average, your GPA.

Okay.

I'm listening.

Think of your overall GPA as your average total cost, ATC.

Think of your grade in the next course you take as your marginal cost, MC.

By marginal grade contribution.

Right.

Now, if your grade in the next course, MC is lower than your current GPA, ATC, what happens to your GPA?

It goes down.

That C plus really hurt.

Exactly.

So if MC is below APC, it pulls the average down.

ATC must be falling.

Makes sense.

Now, what if your grade in the next course, MC, is higher than your current GPA?

ATC.

Then my GPA goes up and A pulls the average up.

Right.

So if MC is above ATC, it pulls the average up.

ATC must be rising.

Okay.

So what happens if my next grade is exactly equal to my GPA?

Your GPA stays exactly the same, right?

It doesn't change.

And this is the key point.

When MC equals ATC, ATC is neither rising nor falling.

It must be at its minimum point, the bottom of the U.

Wow.

So the marginal cost curve has to intersect the average total cost curve exactly at the minimum point of the ATC curve.

Always.

And it intersects from below because MC was below ATC, pulling it down, and then rises above ATC, pulling it up.

This relationship is vital for understanding firm decisions and market supply.

That GPA analogy really helps lock it in.

It's a good one.

And this MCATC logic has real world consequences.

Think about electricity pricing.

Most of us pay for electricity based on the average cost to produce it over a month or whatever.

Yeah, seems normal.

But the marginal cost of generating electricity can swing wildly.

It's much, much higher during peak demand hours, like a hot summer afternoon when everyone's AC is blasting.

Right.

They have to fire up extra less efficient power plants.

Exactly.

But because we consumers usually don't see that high marginal cost, we don't have a strong incentive to cut back during those peak times.

Our demand is inefficiently high when costs are highest.

I see.

We're responding to the average, not the margin.

Precisely.

And that's where things like smart grid technology come in.

The idea is to show consumers something closer to the real time marginal cost.

So you might choose to run your dishwasher late at night when the marginal cost, and thus the price, is lower.

Exactly that.

Using marginal cost pricing to encourage more efficient consumption patterns.

It's a direct application of these cost curve concepts.

Very cool.

Okay, so all this cost stuff, fixed variable marginal average, has mostly been about the short run, right?

Where land was fixed.

Primarily, yes.

The U -shaped ATC curve and the rising MC curve are characteristic of short run analysis where at least one input is fixed, leading to diminishing returns.

But you said in the long run, everything is variable.

Firms can change anything.

That's the key difference.

In the long run, the firm can adjust all its inputs.

It can choose its amount of land, its core technology.

What was a fixed cost in the short run becomes a choice variable in the long run.

So Selena's Gourmet Salsas, your example, they could decide to buy way more fancy automated salsa making machines.

Exactly.

That would mean a higher fixed cost paying for the machines, but maybe it lowers their variable cost per jar, less labor needed.

Or they could stick with less equipment, lower fixed costs, but maybe need more workers.

So higher variable costs.

Right.

It's a trade off.

And because they can choose their level of fixed cost in the long run, there isn't just one short run average total cost curve anymore.

There are many possible short run average total cost SRATC curves, one for each possible level of fixed cost the firm might choose.

Okay.

A whole family of U -shaped SRDC curves.

Then what's the long run curve?

The long run average total cost LRATC basically traces out the lowest possible average cost for each level of output.

Assuming the firm has had time to choose the optimal level of fixed cost for that specific output.

So visually, how does that look?

Imagine all those U -shaped SRATC curves for different factory sizes or equipment levels.

The LRATC curve is like a broader, shallower U -shape that just touches the bottom most point of each relevant SRTC curve.

It sort of envelops them from below.

Okay.

It represents the best possible average cost achievable for any output level given time to adjust everything.

Perfect summary.

So if a firm expects a small temporary increase in demand, it might just produce more using its existing setup, moving along its current SRTC curve, probably at a higher average cost.

But if they expect demand to stay high permanently.

Then in the long run, they'll invest in a bigger factory or more equipment, shifting to a different SRATC curve that allows them to produce that higher output at the lowest possible cost, touching the LRATC curve at that new higher output level.

So the LRATC guides the long -term investment decisions about scale.

Exactly.

And the shape of that LRATC curve tells us about returns to scale.

Okay.

Another returns concept.

How is this different from diminishing returns to an input?

Good question.

Diminishing returns was about adding one variable input while holding others fixed.

Short run.

Returns to scale is about what happens to average cost when you increase all inputs proportionally.

Long run.

Ah, scaling up the whole operation.

Right.

If the LRATC curve is sloping downward, meaning average costs fall as the firm gets bigger, we call that increasing returns to scale, or more commonly, economies of scale.

Bigger is cheaper.

Per unit.

Why does that happen?

Several reasons.

Big one is specialization.

In larger firms, workers can specialize in narrow tasks and get really, really good at them.

Think assembly lines.

Okay.

Also, large setup costs.

Building a giant car factory costs a fortune.

But if you spread that over millions of cars, the cost per car is lower than for a small workshop.

Makes sense.

And sometimes network externalities, especially in tech, like the more people use Facebook or Windows, the more valuable it becomes to the next user, which helps large networks grow and lower average costs per user served.

So lots of reasons for economies of scale.

What about the other way?

Can firms get too big?

They certainly can.

If the LRATC curve starts sloping upward, meaning average costs increase as the firm gets even bigger, that's decreasing returns to scale or diseconomies of scale.

Why would costs go up if you get bigger?

Usually boils down to problems of coordination and communication.

Imagine a massive sprawling bureaucracy.

It can become slow, inefficient, hard to manage.

Information doesn't flow well.

Decision making gets bogged down.

Too much middle management, maybe?

Could be.

Essentially, the organization becomes unwieldy.

And I guess it could be flat, too.

Yep.

If the LRATC is flat, that's constant returns to scale.

Doubling all your inputs exactly doubles your output, so the average cost per unit stays the same.

Okay.

Increasing, decreasing, constant returns to scale.

Describe the LRATC shape.

Any modern examples?

No.

A fantastic one is the sharing economy.

Think about platforms like Turo for cars or Airbnb for rooms.

How does that relate to costs?

It fundamentally changes the nature of costs.

Owning a car involves a huge fixed cost.

Buying it, insuring it, etc.

Even if you barely drive it.

Right.

It just sits there costing money.

But if you use Turo, you pay a variable cost, the rental fee, only when you actually use the car.

The platform allows individuals to convert what would be a large, personal fixed cost into a pay -as -you -go variable cost.

Ah.

So Karenna, who only needs a car sometimes, avoids the big fixed cost.

Exactly.

And think bigger picture.

If more people share cars via Turo, society needs fewer cars overall to provide the same number of trips.

That massively reduces the total fixed cost tied up in vehicles across the economy.

It's much more efficient resource use.

That's a great example.

Using platforms to essentially lower the economy -wide fixed cost for certain activities.

It is.

And you see the flip side with someone like Amazon investing billions in robotic playhouses.

That's a massive increase in fixed costs.

But presumably it lowers their variable costs.

Dramatically.

Fewer human pickers needed denser storage, faster fulfillment.

They're deliberately choosing high fixed costs to achieve incredibly low variable costs and leverage massive economies of scale for speed and efficiency.

It's all about managing that long run cost structure.

Okay.

Wow.

We've covered a lot of ground here.

From basic production to all these different cost curves and long run decisions.

It's the core of how economists think about firms and supply.

So for someone listening, maybe a student trying to get their head around this, what's the main takeaway?

I'd say focus on the relationships between the concepts and the curves.

Don't just memorize definitions.

Understand why MC intersects ATC at the minimum.

Understand why diminishing returns leads to rising MC.

Visualize it.

Yes.

Draw the curves.

See how a change in fixed cost shifts ATC but not MC.

See how a change in variable input prices shifts MC and ATC.

Practice interpreting them.

Connect it to the real world like we've tried to do.

Definitely.

Look around you.

Why does your local coffee shop have two baristas during the morning rush, but maybe only one later?

That's a marginal decision based on costs and expected output.

Why do huge companies like Amazon exist, economies of scale?

These concepts aren't just textbook diagrams.

They explain the business world.

So thinking about these core economic levers,

how might that change how you, the listener, view the economy?

It gives you a much deeper understanding of business strategy.

Why firms make the choices they do, how markets work, and even the impact of government policies.

It's a powerful lens.

Well, thank you for guiding us through that.

It really clarifies things.

My pleasure.

It's fascinating stuff.

Thank you for joining us on this deep dive into the world of microeconomic costs.

We hope this has been an illuminating session, giving you some valuable tools for your studies or just for understanding the world.

From the Dopp Dive team, thank you for listening.