4.3 — Pricing Strategies

# 4.3 — Pricing Strategies
## ECON 306 · Microeconomic Analysis · Spring 2020
### Ryan Safner Assistant Professor of Economics <a href="mailto:safner@hood.edu"> safner@hood.edu</a> <a href="https://github.com/ryansafner/microS20"> ryansafner/microS20</a> <a href="https://microS20.classes.ryansafner.com"> microS20.classes.ryansafner.com</a>

---

### [1st-Degree Price Discrimination](#14)

### [3rd-Degree Price Discrimination](#19)

### [2nd-Degree Price Discrimination](#34)

### [Is Price Discrimination Good or Bad?](#36)

### [Tying and Bundling](#47)

---

# Profit-Seeking Firms

]

- Any firm with market power seeks to maximize profits

- Wants to (1st) **create** a surplus

]

---

# Profit-Seeking Firms

]

- Any firm with market power seeks to maximize profits

- Wants to (1st) **create** a surplus .hi-purple[and then *extract* some of it as profit]
 - i.e. convert .hi-blue[CS] `$\rightarrow$` .hi-green[`\$\pi\$`]

- Consumers are *still* better off than without the firm because it creates value (.blue[consumer surplus])
  - Just not as *best*-off as under perfect competition

]

---

# Most Firms Create More Value than They Can Capture!

.center[
![:scale 70%](https://www.dropbox.com/s/10htmt0dupcb3kb/nordhaus.jpg?raw=1)
.smallest[
William Nordhaus

(1941-)

Economics Nobel 2018
]
]
]

> “We conclude that [about 2.2%] of the social returns from technological advances over the 1948-2001 period was captured by producers, indicating that most of the benefits of technological change are passed on to consumers rather than captured by producers,” (p.1)

]

.source[Nordhaus, William, 2004, ["Schumpeterian Profits in the American Economy: Theory and Measurement,"](https://www.nber.org/papers/w10433) *NBER Working Paper* 10433]

---

# Price Discrimination

- The most obvious way to capture more surplus is to raise prices
  - But **Law of Demand** `$\implies$` this would turn many customers away!

- Instead, if firm could charge **different** customers with *different WTP* **different** prices for **the same goods**, firm could convert more .blue[consumer surplus] into .green[profit]

- .hi[“Price discrimination”] or .hi[“Variable pricing”]

]

]

---

# The Economics of Pricing Strategy I

- Two conditions are required for a firm to engage in variable pricing:

.pull-right[
.center[
![:scale 70%](https://www.dropbox.com/s/3h37ge5bmvety7y/marketpower.jpg?raw=1)
]
]

---

# The Economics of Pricing Strategy I

- Two conditions are required for a firm to engage in variable pricing:

.hi-purple[2) Firms must be able to prevent resale or arbitrage]
  - Clever customers buy in your lower-price market to resell it in your higher-price market

]

![:scale 70%](https://www.dropbox.com/s/ftej3jleki01d3u/notforresale.jpg?raw=1)
]
]

---

# The Economics of Pricing Strategy II

- Firm *must acquire information* about the variations in its customers' demands

- Can the firm identify consumers' demands **before** they buy the product?

]

---

# The Economics of Pricing Strategy III

(Goolsbee et al., 2013: 397)
]

---

# The Economics of Pricing Strategy IV

- With **perfect information** `$\implies$` .hi[Perfect] or .hi[1st-degree price discrimination]

- .hi-purple[Charge a different price to each customer] (their max WTP)

]

.pull-right[
.center[
![](https://www.dropbox.com/s/jm38urzuu0ffuqp/pricediscrimination.jpeg?raw=1)
]
]

---

# The Economics of Pricing Strategy V

- With **imperfect information** `$\implies$` .hi[3rd-degree price discrimination]

- Separate customers into groups (by demand differences) and charge each group a different price

]

.pull-right[
.center[
![](https://www.dropbox.com/s/jm38urzuu0ffuqp/pricediscrimination.jpeg?raw=1)
]
]

---

# The Economics of Pricing Strategy VI

- .hi[2nd-degree price discrimination]: More **indirect** forms of pricing: tying, bundling, quantity-discounts
 - Firm does **not** have enough information to categorize customers into groups
 - Consumers **self-select** into their own group

]

.pull-right[
.center[
![](https://www.dropbox.com/s/jmnc0zx472fzcki/2019holidaydiscount.png?raw=1)
]
]

---

# 1st-Degree Price Discrimination

---

# 1st-Degree Price Discrimination I

]

- If firm has *perfect information* about every customer's demand before purchase:

- .hi[Perfect] or .hi[1st-degree price discrimination]: firm charges *each* customer their maximum willingness to pay
 - “walks” down the market demand curve customer by customer

]

---

# 1st-Degree Price Discrimination II

]

- Firm converts *all* consumer surplus into profit!

- Produces the competitive amount `$(q_c)$`!

]

---

# 1st-Degree Price Discrimination: Example

.pull-left[
.center[
![:scale 100%](https://www.dropbox.com/s/hquylrhjqfsuuhj/collegepd.png?raw=1)
]
]

.pull-right[
.center[
![:scale 100%](https://www.dropbox.com/s/43hfqc3896n8z5l/collegecosts.jpg?raw=1)
]
]

---

# Big Data and Perfect Price Discrimination

---

# 3rd-Degree Price Discrimination

---

# 3rd-Degree Price Discrimination I

- Firms almost never have perfect information about their customers

- But they can often separate customers by .hi-purple[observable characteristics] into .hi-purple[different groups] with similar demands *before purchasing*

]

---

# 3rd-Degree Price Discrimination I

- Firms .hi[segment] the market or engage in .hi[3rd-degree price discrimination] by charging different prices to different *groups* of customers

- By far the most common type of price-discrimination
]

---

# 3rd-Degree Price Discrimination II

.center[
.smallest[
Business Travelers (Less Elastic)
]
]
<img src="4.3-slides_files/figure-html/unnamed-chunk-4-1.png" style="display: block; margin: auto;" />
]

<img src="4.3-slides_files/figure-html/unnamed-chunk-5-1.png" style="display: block; margin: auto;" />
]
.smallest[
Consider airlines: different groups of travelers have different demands & price elasticities
]
---

# 3rd-Degree Price Discrimination II

.center[
.smallest[
Business Travelers (Less Elastic)
]
]
<img src="4.3-slides_files/figure-html/unnamed-chunk-6-1.png" style="display: block; margin: auto;" />
]

.center[
.smallest[
Vacationers (More Elastic)
]
]
<img src="4.3-slides_files/figure-html/unnamed-chunk-7-1.png" style="display: block; margin: auto;" />
]
.smallest[
The firm could charge a **single price** to all travelers and earn some .hi-green[profit]
]
---

# 3rd-Degree Price Discrimination II

.center[
.smallest[
Business Travelers (Less Elastic)
]
]
<img src="4.3-slides_files/figure-html/unnamed-chunk-8-1.png" style="display: block; margin: auto;" />
]

.center[
.smallest[
Vacationers (More Elastic)
]
]
<img src="4.3-slides_files/figure-html/unnamed-chunk-9-1.png" style="display: block; margin: auto;" />
]

.smallest[
With **different prices**: raise price on inelastic travelers, lower price on elastic travelers, earn .hi-green[*more* profit]! 
]

---

# 3rd-Degree Price Discrimination: Examples I

---

# 3rd-Degree Price Discrimination: Examples II

---

# 3rd-Degree Price Discrimination: Examples III

---

# 3rd-Degree Price Discrimination: Examples IV

.pull-left[
.center[
![:scale 100%](https://www.dropbox.com/s/3klmng6p3aetahm/gothardcover.jpg?raw=1)
]
]

.pull-right[
.center[
![:scale 100%](https://www.dropbox.com/s/0pl4wf7my61ah7h/gotpaperback.jpg?raw=1)
]
]

---

# Pricing and Markup

- Firm treats each segment as a *different* market
  1. Find q*: `$MR(q)=MC(q)$` 
  2. Raise p* to maximum WTP (Demand)

- Lerner index implies optimal markup for each segment, again:
`$$\underbrace{\frac{p-MC(q)}{p}}_{\text{Markup % of Price}}=-\frac{1}{\epsilon}$$`
]
]

# 3rd-Degree Price Discrimination: Numerical Example

.content-box-green[
.green[**Example**]: Suppose you run a bar in downtown Frederick, and estimate the nightly demands for beer from undergraduates `$(U)$` and graduates `$(G)$` to be:

`$$\begin{align*}
q_U&=18-4p_U\\
q_G&=12-p_G\\
\end{align*}$$`

Assume the only cost of producing a beer is a constant marginal (and average) cost of $2.

]

1. If your bar could not price discriminate, how much profit would the bar earn?

2. If you could price discriminate, how much profit would the bar earn?

---

# Ways to Segment Markets

- Past purchase behavior
  - repeat customers (more price sensitive)

- By location
  - local demand characteristics
]

.pull-right[
.center[
![](https://www.dropbox.com/s/5l6vjvkao8q4amp/variablepricing.png?raw=1)
]
]
---

# 2nd-Degree Price Discrimination

---

# 2nd-Degree Price Discrimination I

]

- If firm *cannot* identify customers' demands or types before purchase

- .hi[Indirect] or .hi[2nd-degree price discrimination]: firm offers difference price-quantity bundles and allows customers **self-select** their offer

- Ex: **quantity-discounts** or **block pricing**
  - Larger quantities offered at lower prices

]

---

# Is Price Discrimination Good or Bad?

---

# Is Price Discrimination Good or Bad? I

- Ideal competitive market, `$q^*$` where `$p^c=MC$`
]

]

---

# Is Price Discrimination Good or Bad? I

- Ideal competitive market, `$q^c$` where `$p^c=MC$`

- A pure monopolist would produce less `$q^m$` at higher `$p^m$`
  - reduce .blue[consumer surplus] and create **deadweight loss**

- Transfer of some surplus from consumers to producers

]

]

---

# Is Price Discrimination Good or Bad? I

- A price-discriminating monopolist transfers MORE surplus from consumers to producers

- But encourages monopolist to produce more than the pure monopoly level and reduce deadweight loss!
  - At best, also produces at competitive output level! 
]

]

---

# Is Price Discrimination Good or Bad? II

- Price-discrimination creates incentives for innovation and risk-taking

- Firms with high fixed costs of investment earn great profits, can recover their fixed costs

- Might not do so without ability to price-discriminate
]

---

# Is Price Discrimination Good or Bad? III

- As with markups in general, price discrimination has everything to do with .hi[price elasticity of demand]

- If you are paying too much and losing consumer surplus, the real "problem" is that .hi-purple[your demand is very inelastic]
  - fewer options, a particular brand, or a necessity, limited time, etc

- If you want to pay less, .hi-purple[buy generic] (more elastic)

]

---

# How to Be a Savvy Consumer

.pull-left[
.quitesmall[
- Realize that any “sales” and “discounts” are calculate to make *the store* more money

- But it *can* make you better off as a consumer too if you are smart

- Think about your .blue[consumer surplus]!

- If you were *already* planning to buy the product, a fall in price is a good deal for you
  - Your demand is less elastic

- If you *weren’t* going to buy the product before, and now you do, the sale was effective for the store, and you likely don’t get much surplus
  - Your demand is more elastic

]
]

---

# Behavioral Economics

---

# Price Discrimination vs. Price Differences

- .hi[Price discrimination] is selling *identical* goods to people at different prices

- But not everytime people pay different prices means it is price discrimination

- Sometimes it is truly different goods that people are paying different prices for
  - If *costs* to firm are *different* for different versions (color, size, etc.), it is a *different* good, *not* price discrimination

]

]

---

# Price Discrimination vs. Price Differences

- .hi-green[Example]: bottled sparkling water often more expensive than Coca Cola
  - Could be because sparkling water drinkers have more elastic demand than Coke drinkers
  - Or could be that it is more expensive to package sparkling water (economies of scale with greater number of Coke drinkers)

]

]

---

# Price Discrimination vs. Price Differences

- The only way to tell the difference is to see what happens if demand changes price elasticity (and costs do not change)
  - Price discrimination requires market power, firm with market power marks up price based on `$\frac{1}{\epsilon}$`
  - Competitive firm only sets `$p=MC$`, so change in elasticity has no effect on price

- See [today’s class notes](/class/4.3-class) for a graphical demonstration
]

]

---

# Tying and Bundling

---

# Tying I

- Firms often .hi[tie] multiple goods together, where you must buy both goods in order to consume the product
  - One good often the "base" and the other are "refills" that you may need to buy more of

- This is actually a method of .hi-purple[*intertemporal* price-discrimination]!

]

.pull-right[
.center[
!["scale 80%](https://www.dropbox.com/s/k7yjp4lqpazmzrh/printerandink.jpg?raw=1)

![:scale 40%](https://www.dropbox.com/s/80b8y4pu4e6q9t0/razorheads.jpg?raw=1)
]

]

---

# Tying II

- Companies often **sell printers at marginal cost** (no markup) and sell the **ink/refills at a much higher markup**

- **Reduce arbitrage**:
  - printer requires specific ink
  - ink only words with that specific printer

]

---

# Tying II

- Segment the market into:

1. .hi-purple[High-volume users]: buy more ink over time; pay more per sheet printed

2. .hi-purple[Low-volume users]: buy less ink; pay less per sheet printed

- **Indirect** price-discrimination: firms **don't know** what kind of user you are in advance

]

---

# Tying: Good or Bad?

- Again, a tradeoff:

- Increased profits and reduced consumer surplus, reduced deadweight loss

- Spreads fixed cost of research & development over more users

]

---
 
 # Tying: Good or Bad?

- If printers & ink were **not** tied:
  - **printers** would be **more expensive**
  - **ink** would be  **cheaper**

- High-volume users would keep buying ink and save money (vs. tied)

- Low-volume users might not buy the (now expensive) printer at all!

]

---

# Bundling I

- Firms often .hi[bundle] products together as a single package, and refuse to offer individual parts of the package

- Often, consumers do not want all products in the bundle

- Or, if they were able to buy just part of the bundle, they would *not* buy the other parts
]

---

# Bundling II

.content-box-green[
.green[**Example**]: Consider two consumers, each have different reservation prices to buy components in Microsoft Office bundle

]

|    | Amy's WTP | Ben's WTP |
|----|-----|-----|
| MS Word | $70 | $40 |
| MS Excel | $50 | $60 |

]

---

# Bundling II

.content-box-green[
.green[**Example**]: Consider two consumers, each have different reservation prices to buy components in Microsoft Office bundle

]

|    | Amy's WTP | Ben's WTP |
|----|-----|-----|
| MS Word | $70 | $40 |
| MS Excel | $50 | $60 |

]

- MS Word: both would buy at $40, generating $80 of revenues
]
]

---

# Bundling II

.content-box-green[
.green[**Example**]: Consider two consumers, each have different reservation prices to buy components in Microsoft Office bundle

]

|    | Amy's WTP | Ben's WTP |
|----|-----|-----|
| MS Word | $70 | $40 |
| MS Excel | $50 | $60 |

]

- MS Word: both would buy at $40, generating $80 of revenues

- MS Excel: both would buy at $50, generating $100 of revenues
]
]

---

# Bundling II

.content-box-green[
.green[**Example**]: Consider two consumers, each have different reservation prices to buy components in Microsoft Office bundle

]

|    | Amy's WTP | Ben's WTP |
|----|-----|-----|
| MS Word | $70 | $40 |
| MS Excel | $50 | $60 |

]

- MS Word: both would buy at $40, generating $80 of revenues

- MS Excel: both would buy at $50, generating $100 of revenues

- Total revenues of individual sales: $180
]

]

---

# Bundling II

.content-box-green[
.green[**Example**]: Consider two consumers, each have different reservation prices to buy components in Microsoft Office bundle

]

|    | Amy's WTP | Ben's WTP |
|----|-----|-----|
| MS Word | $70 | $40 |
| MS Excel | $50 | $60 |
| Bundle | $120 | $100 |

]

- MS Word: both would buy at $40, generating $80 of revenues

- MS Excel: both would buy at $50, generating $100 of revenues

- Total revenues of individual sales: $180

- Microsoft can instead add their individual reservation prices and bundle products together to force both consumers to buy both products

- .hi-purple[Bundle]: both buy at $100, generating $200 revenue
]
]

---

# Bundling: Good or Bad?

- Again, a tradeoff:

- Increased profits and reduced consumer surplus, reduced deadweight loss

- Spreads fixed cost of research & development over more users

- Goods with high fixed costs and low marginal costs (software, TV, music) increase profits from bundling
  - increases innovation and investment in these industries

]