Beeronometrics

class: center, middle, inverse, title-slide

.title[
# Beeronometrics
]
.subtitle[
## Identifying the effect of market concentration level in beer consumption
]
.author[
### Professor: Cristoph Hanck <br> <br> Camilo Santa Cruz <br> 235633
]
.date[
### 1 August. 2023
]

---

# Motivation

.content-box-blue[- Competition historically recognized as a key factor for market performance: product quality, innovation, efficiency levels and consumer welfare.
]
<br>
.content-box-grey[- Given this, the present study seeks to identify the relation between consumption of beer and competition in beer market for 96 countries.
]

<br>
.content-box-blue[- Is there a relation between market concentration and consumption per capita? (market efficiency).
]

---

# Theoretical framework

<a name=cite-OXFORD2022></a>([Oxford-Economics, 2022](#bib-OXFORD2022)),  <a name=cite-Swinnen2011></a>([Swinnen, 2011](#bib-Swinnen2011)),  <a name=cite-BREWERS2017></a>([Europe-Economics, 2017](#bib-BREWERS2017)),  <a name=cite-RADCLIFFE2021></a>([Radcliffe, 2021](#bib-RADCLIFFE2021)),  <a name=cite-PARSONS2013></a>([Parsons and de
Vanssay, 2013](#bib-PARSONS2013)):

.content-box-blue[- Beer economics: Factors that influence production, sales and consumption of beer
]

.content-box-grey[- Economics of scale in beer markets
]

<a name=cite-Cournot2012></a>([Cournot and Ligaran, 2012](#bib-Cournot2012)), <a name=cite-Maddala></a>([Maddala and Miller, 2004](#bib-Maddala)), <a name=cite-Alhabeeb2012></a>([Alhabeeb and Moffitt, 2012](#bib-Alhabeeb2012)), <a name=cite-KEAT2012></a>([Keat, Young, and Erfle, 2012](#bib-KEAT2012)):

.content-box-blue[- Determinants of supply and demand: Inputs costs, Price, Income, Quantity that firms sold and produce and Consumer behavior in market.
]

.content-box-grey[- Industrial economics, Business Economics and Market concentration:
Levels of output under monopoly and competition.
]
---

# Economies of scale and efficiency

.pull-left[
`\begin{align}
Q = \beta_1 Q^2 - \beta_2 Q + \beta_3 + \epsilon
\end{align}`
<img src="data:image/png;base64,#BeeronometricsPresentation_files/figure-html/myfig2-1.png" width="450px" style="display: block; margin: auto;" />
]

.pull-right[
`\begin{align}
Q = \alpha - \beta P + \epsilon \\
\text{Under competition:} \ P = MC
\end{align}`

<img src="data:image/png;base64,#BeeronometricsPresentation_files/figure-html/efficiency-1.png" width="450px" style="display: block; margin: auto;" />
]
---

# Theoretical shifts of demand and supply in market

.pull-left[
`\(Q = \alpha + \beta_1 P + \beta_2 IHH + \epsilon\)` 
<img src="data:image/png;base64,#BeeronometricsPresentation_files/figure-html/shifts-1.png" width="450px" style="display: block; margin: auto;" />
]

.pull-right[
`\(Q = \alpha - \beta_1 P + \beta_2 IHH + \beta_3 Income + \epsilon\)`

<img src="data:image/png;base64,#BeeronometricsPresentation_files/figure-html/shifts1-1.png" width="450px" style="display: block; margin: auto;" />
]

---

# Empirical evidence on beer market

Beer market appears to follow characteristics of normal goods: negative price - consumption relation or price - sales relation from the optic of firms
.content-box-grey[<a name=cite-Rumankova2018></a>([Sanova and Kolar, 2018](#bib-Rumankova2018)): Negative price - elasticities and positive income elasticities (beer follow normal good characteristics) (-0.64).

]

.content-box-blue[<a name=cite-COLEN2010></a>([Colen and Swinnen, 2010](#bib-COLEN2010)): negative price - consumption elasticity identified.

]

.content-box-grey[<a name=cite-NELSON2014></a>([Nelson, 2014](#bib-NELSON2014)): Meta - analysis gives evidence of a price elasticity of -0.34%.

]

.content-box-blue[<a name=cite-Strjer2014></a>([Madsen and Wu, 2014](#bib-Strjer2014)): Give evidence on scale economics. An increase in sales is translated in an less than proportional increase in different type of costs.

]

.content-box-grey[<a name=cite-Tremblay1985></a>([Tremblay, 1985](#bib-Tremblay1985)): Suggest a quadratic relation between concentration and price. For his study detects no evidence that increase in concentration leads to increase in prices.

]

---

.content-box-blue[<a name=cite-Orley2015></a>([Ashenfelter, Hosken, and Weinberg, 2015](#bib-Orley2015)): Find that concentration (because of merging) is translated in price increases of 2% and efficiency effect (because of economies of scale) is translated in a delayed price decrease of -1.8.

]

.content-box-blue[<a name=cite-LYNK1984></a>([Lynk, 1984](#bib-LYNK1984)): Rising concentration is translated into more beer sales. The shift in sales from small to large brewers is accomplished not by interference with the supply of the small, but by expansion of supply from the large.

]

.content-box-grey[<a name=cite-Miller2016></a>([Miller and Weinber, 2016](#bib-Miller2016)): Detect with difference in difference methods that concentration (because big firm merging) is translated in price increase.

]

.content-box-blue[<a name=cite-Polemis2020></a>([Polemis, Stengos, and Tzeremes, 2020](#bib-Polemis2020)): Quadratic effect of concentration in manufacturing market in USA.

]

.content-box-grey[<a name=cite-Salas2019></a>([Salas-Velasco, 2019](#bib-Salas2019)): Production per capita explained by competition, competitiveness and innovation. No competition effects found when explaining production per capita. (competition maybe is endogenous).

]

---

# Empirical strategy

.content-box-blue[Panel data methods: 96 countries - 10 years

]

.content-box-grey[Data from: EUROMONITOR INTERNATIONAL, WORLD BANK, FAO

]

.content-box-blue[Variables:

- Beer (liters per 1000 inhabitants): `\(\frac{Q_{it}}{N_{it}/1000}\)`

- Market concentration (HHI): `\(\sum s_{fit}^2\)` - Max value HHI: 100 (Company level)
]

.content-box-grey[
- Price (dollars per 1000 liters): `\(P_{it} \equiv \frac{R_{it}}{Q_{it}} \equiv \frac{P_{it}\ast Q_{it}}{Q_{it}}\)`

- Income (Disposable dollars per 1000 inhabitants): `\(\frac{I_{it}}{N_{it}/1000}\)`

- Others (Market share of microfirms that Euromonitor take as residual account)

- Wine consumption (dollars per 1000 liters): `\(\frac{Qw_{it}}{N_{it}/1000}\)`

]
---

# Packages

readxl: Import excel files into R <a name=cite-Wickham2023></a>([Wickham and Bryan, 2023](#bib-Wickham2023))

dplyr and tidyverse: Processing the data easily. <a name=cite-Wickham2019></a>([Wickham, Averick, Bryan, Chang, McGowan, François, Grolemund, Hayes, Henry, Hester, Kuhn, Pedersen, Miller, Bache, Müller, Ooms, Robinson, Seidel, Spinu, Takahashi, Vaughan, Wilke, Woo, and Yutani, 2019](#bib-Wickham2019)), <a name=cite-HadleyWickham2023></a>([Wickham, François, Henry, Müller, and Vaughan, 2023](#bib-HadleyWickham2023))

ggplot: Representation of the series and its distributions. <a name=cite-Wickham2016></a>([Wickham, 2016](#bib-Wickham2016))

plm: Running the panel model regressions. <a name=cite-Croissant2018></a>([Croissant and Millo, 2018](#bib-Croissant2018))

stargazer: Preparing the table of regressions and exporting easily to HTML and latex format. <a name=cite-Marek2022></a>([Hlavac, 2022](#bib-Marek2022))

---

# Data processing

.content-box-blue-[Data processing:

- RStudio - Packages: readxl, tidyverse, dplyr, plm, ggplot, stargazer, kable
]

.font25[
.pull-left[

```r
names <- colnames(Income)[2:104] # Taking only countries (103 countries)
Income1 <- Income %>% pivot_longer(cols = contains(names), # Making data longer with tidyverse
  names_to = "Country",
  values_to = "Income") %>% group_by(Country)
```
]]

.font25[
.pull-right[
<table>
 <thead>
  <tr>
   <th style="text-align:left;"> Ano </th>
   <th style="text-align:right;"> Dominican Republic </th>
   <th style="text-align:right;"> Ecuador </th>
   <th style="text-align:right;"> El Salvador </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:left;"> 2017 </td>
   <td style="text-align:right;"> 63361.0 </td>
   <td style="text-align:right;"> 72146.4 </td>
   <td style="text-align:right;"> 29573.3 </td>
  </tr>
  <tr>
   <td style="text-align:left;"> 2018 </td>
   <td style="text-align:right;"> 66837.0 </td>
   <td style="text-align:right;"> 73343.1 </td>
   <td style="text-align:right;"> 30523.4 </td>
  </tr>
  <tr>
   <td style="text-align:left;"> 2019 </td>
   <td style="text-align:right;"> 70101.3 </td>
   <td style="text-align:right;"> 73079.0 </td>
   <td style="text-align:right;"> 31685.4 </td>
  </tr>
</tbody>
</table>
<br>
<br>
<table>
 <thead>
  <tr>
   <th style="text-align:left;"> Year </th>
   <th style="text-align:left;"> Country </th>
   <th style="text-align:right;"> Income </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:left;"> 2020 </td>
   <td style="text-align:left;"> India </td>
   <td style="text-align:right;"> 2220028.9 </td>
  </tr>
  <tr>
   <td style="text-align:left;"> 2020 </td>
   <td style="text-align:left;"> Indonesia </td>
   <td style="text-align:right;"> 674303.2 </td>
  </tr>
  <tr>
   <td style="text-align:left;"> 2020 </td>
   <td style="text-align:left;"> Japan </td>
   <td style="text-align:right;"> 2858176.9 </td>
  </tr>
</tbody>
</table>
]]
---

# Series plots

```r
Beer_HII <- ggplot(Beer4, aes(x = log(IHHA1)),na.rm = TRUE) + geom_histogram(aes(y = after_stat(density)), binwidth = .5, colour = "darkblue",na.rm = TRUE, fill = 0.3) + geom_density(alpha = .7, fill= "#F1952D",na.rm = TRUE) + xlab("HHI") +  theme(axis.text = element_text(size = 70)) + theme_classic()
Beer_HII
```

<img src="data:image/png;base64,#BeeronometricsPresentation_files/figure-html/myfig-1.png" width="450px" style="display: block; margin: auto;" />
---

```r
Beer_HII1 <- ggplot(data = Beer4, aes(y = log(IHHA1), x = Observation),na.rm = TRUE) + xlim (1,960) +  geom_line(col = "darkblue",na.rm = TRUE, lwd = 1) + 
  ylab("ln(HHI)") + labs(title = "HHI Index", col = "darkblue") + theme(axis.text = element_text(size = 100)) + theme_classic()
Beer_HII1
```

---
class: slide1
background-color: #004c93

# Results

---

# Correlations

```r
Correlation <- Beer7 %>% dplyr::select("LQ_Capita", "LPrice","LIncome_Capita", "LIHHA1","LPrice_Wine", "LLitres_Wine", "LOthers")
M <- cor(Correlation, use='complete.obs')
corrplot1 <- corrplot(M, type="upper","color",diag=FALSE, addCoef.col = "black", addgrid.col = "black", tl.col = "black", col = brewer.pal(n = 6, name="Reds"))
```

<img src="data:image/png;base64,#BeeronometricsPresentation_files/figure-html/myfigure2-1.png" width="450px" style="display: block; margin: auto;" />
---

# Coding of Panel Data Models

`\begin{align}
\log(q)& =\ \beta_0\ +\ \beta_1(HHI)\  +\ \beta_2(Others)\\
\log(q)& =\ \beta_0\ +\ \beta_1(HHI)\ +\ \beta_2\log(Price)\ +\ \beta_3log(Income)\ +\ \beta_4(Others)\\
\log(q)& =\ \beta_0\ +\ \beta_1(HHI)\ +\ \beta_2\log(Price)\ +\ \beta_3log(Income)
\end{align}`

.font25[
.pull-left[

```r
pooling1 <- plm(LQ_Capita ~ LIHHA1 + log(Others), data = Beer4,na.action = na.omit, index=c("Country", "Year"), model="pooling")
fe1 <- plm(LQ_Capita ~ LIHHA1 + log(Others), data = Beer4,na.action = na.omit, index=c("Country", "Year"), model="within")
re1 <- plm(LQ_Capita ~ LIHHA1 + log(Others), data = Beer4,na.action = na.omit, index=c("Country", "Year"), model="random")
fd1 <- plm(LQ_Capita ~ LIHHA1 + log(Others), data = Beer4,na.action = na.omit, index=c("Country", "Year"), model="fd")
be1 <- plm(LQ_Capita ~ LIHHA1 + log(Others), data = Beer4,na.action = na.omit, index=c("Country", "Year"), model="between")
```

```r
fe2 <- plm(LQ_Capita ~ LIHHA1 + LPrice + LIncome_Capita + log(Others), data = Beer4,na.action = na.omit, index=c("Country","Year"), model="within")
re2 <- plm(LQ_Capita ~ LIHHA1 + LPrice + LIncome_Capita + log(Others) , data = Beer4,na.action = na.omit, index=c("Country","Year"), model="random")
fe3 <- plm(LQ_Capita ~  LIHHA1 + LPrice + LIncome_Capita + log(Others) + LLitres_Wine, data = Beer7,na.action = na.omit, index=c("Country","Year"), model="within")
re3 <- plm(LQ_Capita ~  LIHHA1 + LPrice + LIncome_Capita + log(Others) + LLitres_Wine, data = Beer7,na.action = na.omit, index=c("Country","Year"), model="random")
```

]]

.font25[
.pull-right[

```r
stargazer1 <- stargazer(pooling2, pooling3, pooling4, type = "html",se = rob_se, no.space = TRUE, column.sep.width = "-15pt", header = F, single.row = T, title = "POLS")

stargazer3 <- stargazer(pooling1, fe1, re1, fd1,be1, type = "html", single.row = T, no.space=TRUE, column.sep.width = "-15pt", header = F, title = "Consumption = f Competition)")

stargazer4 <- stargazer(fe2, fe3, re2, re3, type = "html", se = rob_se2, no.space=TRUE, single.row = T, column.sep.width = "-15pt", row.sep.width = -5,header = F, title = "FE and RE")
```

```r
rob_se2 <- list(sqrt(diag(vcovHC(fe2, type ="HC1", method = "white1"))),
               sqrt(diag(vcovHC(fe3, type = "HC1", method = "white1"))),
               sqrt(diag(vcovHC(re2, type = "HC1", method = "white1"))),
               sqrt(diag(vcovHC(re3, type = "HC1", method = "white1")))
```
]]

---
class: regression

### POLS models

<table style="text-align:center"><caption><strong>POLS</strong></caption>
<tr><td colspan="4" style="border-bottom: 1px solid black"></td></tr><tr><td style="text-align:left"></td><td colspan="3"><em>Dependent variable:</em></td></tr>
<tr><td></td><td colspan="3" style="border-bottom: 1px solid black"></td></tr>
<tr><td style="text-align:left"></td><td colspan="3">LQ_Capita</td></tr>
<tr><td style="text-align:left"></td><td>(1)</td><td>(2)</td><td>(3)</td></tr>
<tr><td colspan="4" style="border-bottom: 1px solid black"></td></tr><tr><td style="text-align:left">LIHHA1</td><td>-1.563<sup>***</sup> (0.224)</td><td>-0.435<sup>**</sup> (0.171)</td><td>-0.258<sup>**</sup> (0.114)</td></tr>
<tr><td style="text-align:left">LPrice</td><td></td><td>-0.482<sup>***</sup> (0.065)</td><td>-0.484<sup>***</sup> (0.064)</td></tr>
<tr><td style="text-align:left">LIncome_Capita</td><td></td><td>0.796<sup>***</sup> (0.047)</td><td>0.804<sup>***</sup> (0.047)</td></tr>
<tr><td style="text-align:left">log(Others)</td><td>-0.180<sup>***</sup> (0.062)</td><td>-0.078 (0.048)</td><td></td></tr>
<tr><td style="text-align:left">Constant</td><td>16.698<sup>***</sup> (0.977)</td><td>8.423<sup>***</sup> (0.852)</td><td>7.505<sup>***</sup> (0.578)</td></tr>
<tr><td colspan="4" style="border-bottom: 1px solid black"></td></tr><tr><td style="text-align:left">Observations</td><td>951</td><td>951</td><td>959</td></tr>
<tr><td style="text-align:left">R<sup>2</sup></td><td>0.079</td><td>0.372</td><td>0.368</td></tr>
<tr><td style="text-align:left">Adjusted R<sup>2</sup></td><td>0.077</td><td>0.369</td><td>0.366</td></tr>
<tr><td style="text-align:left">F Statistic</td><td>40.424<sup>***</sup> (df = 2; 948)</td><td>139.977<sup>***</sup> (df = 4; 946)</td><td>185.115<sup>***</sup> (df = 3; 955)</td></tr>
<tr><td colspan="4" style="border-bottom: 1px solid black"></td></tr><tr><td style="text-align:left"><em>Note:</em></td><td colspan="3" style="text-align:right"><sup>*</sup>p<0.1; <sup>**</sup>p<0.05; <sup>***</sup>p<0.01</td></tr>
</table>

---
class: regression

### Simple panel regression

<table style="text-align:center"><caption><strong>Consumption = f (Competition)</strong></caption>
<tr><td colspan="6" style="border-bottom: 1px solid black"></td></tr><tr><td style="text-align:left"></td><td colspan="5"><em>Dependent variable:</em></td></tr>
<tr><td></td><td colspan="5" style="border-bottom: 1px solid black"></td></tr>
<tr><td style="text-align:left"></td><td colspan="5">LQ_Capita</td></tr>
<tr><td style="text-align:left"></td><td>(1)</td><td>(2)</td><td>(3)</td><td>(4)</td><td>(5)</td></tr>
<tr><td colspan="6" style="border-bottom: 1px solid black"></td></tr><tr><td style="text-align:left">LIHHA1</td><td>-1.563<sup>***</sup> (0.177)</td><td>-0.230<sup>***</sup> (0.066)</td><td>-0.249<sup>***</sup> (0.066)</td><td>-0.059 (0.054)</td><td>-1.684<sup>***</sup> (0.593)</td></tr>
<tr><td style="text-align:left">log(Others)</td><td>-0.180<sup>***</sup> (0.052)</td><td>-0.018 (0.011)</td><td>-0.019<sup>*</sup> (0.011)</td><td>0.015 (0.009)</td><td>-0.221 (0.180)</td></tr>
<tr><td style="text-align:left">Constant</td><td>16.698<sup>***</sup> (0.775)</td><td></td><td>11.109<sup>***</sup> (0.307)</td><td>-0.001 (0.003)</td><td>17.262<sup>***</sup> (2.607)</td></tr>
<tr><td colspan="6" style="border-bottom: 1px solid black"></td></tr><tr><td style="text-align:left">Observations</td><td>951</td><td>951</td><td>951</td><td>855</td><td>96</td></tr>
<tr><td style="text-align:left">R<sup>2</sup></td><td>0.079</td><td>0.014</td><td>0.017</td><td>0.007</td><td>0.083</td></tr>
<tr><td style="text-align:left">Adjusted R<sup>2</sup></td><td>0.077</td><td>-0.098</td><td>0.015</td><td>0.004</td><td>0.063</td></tr>
<tr><td style="text-align:left">F Statistic</td><td>40.424<sup>***</sup> (df = 2; 948)</td><td>6.157<sup>***</sup> (df = 2; 853)</td><td>14.562<sup>***</sup></td><td>2.870<sup>*</sup> (df = 2; 852)</td><td>4.216<sup>**</sup> (df = 2; 93)</td></tr>
<tr><td colspan="6" style="border-bottom: 1px solid black"></td></tr><tr><td style="text-align:left"><em>Note:</em></td><td colspan="5" style="text-align:right"><sup>*</sup>p<0.1; <sup>**</sup>p<0.05; <sup>***</sup>p<0.01</td></tr>
</table>

---
class: regression

### FE and RE models

<table style="text-align:center"><caption><strong>FE and RE</strong></caption>
<tr><td colspan="5" style="border-bottom: 1px solid black"></td></tr><tr><td style="text-align:left"></td><td colspan="4"><em>Dependent variable:</em></td></tr>
<tr><td></td><td colspan="4" style="border-bottom: 1px solid black"></td></tr>
<tr><td style="text-align:left"></td><td colspan="4">LQ_Capita</td></tr>
<tr><td style="text-align:left"></td><td>(1)</td><td>(2)</td><td>(3)</td><td>(4)</td></tr>
<tr><td colspan="5" style="border-bottom: 1px solid black"></td></tr><tr><td style="text-align:left">LIHHA1</td><td>-0.150 (0.093)</td><td>-0.116 (0.078)</td><td>-0.157*<sup></sup> (0.092)</td><td>-0.103 (0.071)</td></tr>
<tr><td style="text-align:left">LPrice</td><td>-0.350<sup>***</sup> (0.037)</td><td>-0.260<sup>***</sup> (0.036)</td><td>-0.355<sup>***</sup> (0.039)</td><td>-0.255<sup>***</sup> (0.036)</td></tr>
<tr><td style="text-align:left">LIncome_Capita</td><td>0.562<sup>***</sup> (0.060)</td><td>0.409<sup>***</sup> (0.057)</td><td>0.593<sup>***</sup> (0.060)</td><td>0.366<sup>***</sup> (0.055)</td></tr>
<tr><td style="text-align:left">log(Others)</td><td>-0.021<sup>*</sup> (0.012)</td><td>-0.002 (0.011)</td><td>-0.022<sup>*</sup> (0.012)</td><td>-0.001 (0.012)</td></tr>
<tr><td style="text-align:left">LLitres_Wine</td><td></td><td>0.315<sup>***</sup> (0.043)</td><td></td><td>0.315<sup>***</sup> (0.039)</td></tr>
<tr><td style="text-align:left">Constant</td><td></td><td></td><td>8.070<sup>***</sup> (0.577)</td><td>4.567<sup>***</sup> (0.711)</td></tr>
<tr><td colspan="5" style="border-bottom: 1px solid black"></td></tr><tr><td style="text-align:left">Observations</td><td>951</td><td>931</td><td>951</td><td>931</td></tr>
<tr><td style="text-align:left">R<sup>2</sup></td><td>0.285</td><td>0.359</td><td>0.292</td><td>0.388</td></tr>
<tr><td style="text-align:left">Adjusted R<sup>2</sup></td><td>0.202</td><td>0.284</td><td>0.289</td><td>0.384</td></tr>
<tr><td style="text-align:left">F Statistic</td><td>84.878<sup>***</sup> (df = 4; 851)</td><td>93.387<sup>***</sup> (df = 5; 832)</td><td>388.217<sup>***</sup></td><td>578.003<sup>***</sup></td></tr>
<tr><td colspan="5" style="border-bottom: 1px solid black"></td></tr><tr><td style="text-align:left"><em>Note:</em></td><td colspan="4" style="text-align:right"><sup>*</sup>p<0.1; <sup>**</sup>p<0.05; <sup>***</sup>p<0.01</td></tr>
</table>
---

# FE vs RE (Model efficiency)

Unique errors `\(U_i\)` are not correlated with regressors

<table class="table" style="width: auto !important; margin-left: auto; margin-right: auto;">
<caption>Hausman test</caption>
 <thead>
  <tr>
   <th style="text-align:left;">   </th>
   <th style="text-align:left;"> Regression </th>
   <th style="text-align:right;"> Statistic </th>
   <th style="text-align:right;"> P.value </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:left;"> chisq </td>
   <td style="text-align:left;"> 1 </td>
   <td style="text-align:right;"> 105.517 </td>
   <td style="text-align:right;"> 0.000 </td>
  </tr>
  <tr>
   <td style="text-align:left;"> chisq1 </td>
   <td style="text-align:left;"> 2 </td>
   <td style="text-align:right;"> 4.895 </td>
   <td style="text-align:right;"> 0.298 </td>
  </tr>
  <tr>
   <td style="text-align:left;"> chisq2 </td>
   <td style="text-align:left;"> 3 </td>
   <td style="text-align:right;"> 6.370 </td>
   <td style="text-align:right;"> 0.272 </td>
  </tr>
</tbody>
</table>

---

# Conclusions

- The main purpose of this study has been to verify if the level of concentration in the beer industry at the company level is a determinant of the level of per capita consumption.

- Through the application of panel data models, it has been possible to verify a negative relationship between market concentration and beer consumption, controlling for variables such as market price (proxy of production costs and willingness to pay) or income.

- In the different functional specifications, a negative link is found between the level of concentration and per capita consumption. When performing the Hausman test, the test result suggests that a model with Random Effects fits better the observed data set.

- When controlling for heteroscedasticity in the residuals using White SE, an effect of market concentration on the level of per capita consumption is still identified, although much more moderate than in the case of the pooled regressions.
---

# Future research lines

- A future line of research is the estimation of a simultaneous model to identify the demand and supply of beer and see the separate influence of these on the pressures on the total amount traded and the price. For this it would be opportune to have more data both on the demand side and on the supply side, such as: inputs such as hops, barley, employment (remunerations), capital (GKF), and cost of transport and storage in breweries.

- Incorporating variables that take into account the entire beer value chain would provide a much more detailed drawing of what are the forces that are behind the competitiveness of the beer sector in each country and if these are related to the productive capacity and the increase in sales and consumption within nations.

- It would also be interesting to build a data panel model that considers the possible spatial correlation between countries, to the extent that it is intuited that this relationship may have an influence on those nations that are traditional in beer consumption and also those that are very efficient producing / exporting beer. Spatial integration would also help to solve the problems associated with the correlation between errors.

---

# Thanks for your attention

# Questions

# Recommendations

---

# References

<a name=bib-Alhabeeb2012></a>[Alhabeeb, M. and L.
Moffitt](#cite-Alhabeeb2012) (2012). _Managerial economics: A
mathematical approach._ John Wiley & Sons, Incorporated.

<a name=bib-Orley2015></a>[Ashenfelter, O., D. Hosken, and M.
Weinberg](#cite-Orley2015) (2015). "Efficiencies brewed: Pricing and
consolidation in the US beer industry". In: _The RAND journal of
Economics_ 46.2.

<a name=bib-COLEN2010></a>[Colen, L. and J. Swinnen](#cite-COLEN2010)
(2010). _Beer Drinking Nations: The Determinants of Global Beer
Consumption_. Discussion paper 270. LICOS.

<a name=bib-Cournot2012></a>[Cournot, A. A. and
Ligaran](#cite-Cournot2012) (2012). _Principes de la Théorie des
Richesses_. Cork, BELGIUM: Ligaran Éditions. ISBN: 9782335043013. URL:
[http://ebookcentral.proquest.com/lib/sidunalibro-ebooks/detail.action?docID=5044406](http://ebookcentral.proquest.com/lib/sidunalibro-ebooks/detail.action?docID=5044406).

<a name=bib-Croissant2018></a>[Croissant, Y. and G.
Millo](#cite-Croissant2018) (2018). "Panel Data Econometrics in R: The
plm Package". In: _Journal of Statistical Software_ 27.2, pp. 1-43.
URL:
[https://cran.r-project.org/web/packages/plm/vignettes/A_plmPackage.html](https://cran.r-project.org/web/packages/plm/vignettes/A_plmPackage.html).

<a name=bib-BREWERS2017></a>[Europe-Economics](#cite-BREWERS2017)
(2017). _The Brewers of Europe Cost comparability study - Final
report_. Brewers Europe. URL:
[https://brewersofeurope.org/site/media-centre/index.php?doc_id=997&class_id=31&detail=true](https://brewersofeurope.org/site/media-centre/index.php?doc_id=997&class_id=31&detail=true).
---
<a name=bib-Marek2022></a>[Hlavac, M.](#cite-Marek2022) (2022).
"Stargazer: Well-Formatted Regression and Summary Statistics Tables. R
package version 5.2.3."

URL:
[https://CRAN.R-project.org/package=stargazer](https://CRAN.R-project.org/package=stargazer).

<a name=bib-KEAT2012></a>[Keat, P. G., P. K. Y. Young, and S. E.
Erfle](#cite-KEAT2012) (2012). "Managerial economics. Economic tools
for today's decision Makers".

<a name=bib-LYNK1984></a>[Lynk, W.](#cite-LYNK1984) (1984).
"Interpreting Rising Concentration: The Case of Beer". In: _The journal
of Business_ 57.1, pp. 43 - 55.

<a name=bib-Maddala></a>[Maddala, G. S. and E. M.
Miller](#cite-Maddala) (2004). _Microeconomics: Theory And
Applications_. McGraw-Hill Education. ISBN: 9780070590557. URL:
[https://books.google.de/books?id=EaWu1gsDDLQC](https://books.google.de/books?id=EaWu1gsDDLQC).

<a name=bib-Strjer2014></a>[Madsen, E. S. and Y. Wu](#cite-Strjer2014)
(2014). _Globalization of Brewing and Economies of Scale_. Aarhus
University.

<a name=bib-Miller2016></a>[Miller, N. H. and M. C.
Weinber](#cite-Miller2016) (2016). _The Market Power Effects of a
Merger: Evidence from the U.S. Brewing Industry_. Seminar paper. London
School of Economics.
---
<a name=bib-NELSON2014></a>[Nelson, J. P.](#cite-NELSON2014) (2014).
"Estimating the price elasticity of beer: Meta-analysis of data with
heterogeneity, dependence, and publication bias". In: _Journal of
Health Economics_ 33, pp. 180-187. ISSN: 0167-6296. DOI:
[https://doi.org/10.1016/j.jhealeco.2013.11.009](https://doi.org/https%3A%2F%2Fdoi.org%2F10.1016%2Fj.jhealeco.2013.11.009).

<a name=bib-OXFORD2022></a>[Oxford-Economics](#cite-OXFORD2022) (2022).
_BEER'S GLOBAL ECONOMIC FOOTPRINT_. Oxford - Economics. URL:
[https://worldwidebrewingalliance.org/wp-content/uploads/2022/02/Oxford-Economics-Study-on-the-Economic-Footprint-of-the-Global-Beer-Industry-1.pdf](https://worldwidebrewingalliance.org/wp-content/uploads/2022/02/Oxford-Economics-Study-on-the-Economic-Footprint-of-the-Global-Beer-Industry-1.pdf).

<a name=bib-PARSONS2013></a>[Parsons, C. R. and X. de
Vanssay](#cite-PARSONS2013) (2013). _MEASURING MARKET POWER IN THE
JAPANESE BEER INDUSTRY_. AAWE WORKING PAPER No. 131. AMERICAN
ASSOCIATION OF WINE ECONOMISTS. URL:
[https://wine-economics.org/wp-content/uploads/2013/01/AAWE_WP131.pdf](https://wine-economics.org/wp-content/uploads/2013/01/AAWE_WP131.pdf).

<a name=bib-Polemis2020></a>[Polemis, M. L., T. Stengos, and N. G.
Tzeremes](#cite-Polemis2020) (2020). "Modeling the effect of
competition on US manufacturing sectors'efficiency: an order-m frontier
analysis". In: _Journal of Productivity Analysis_ 54.1, pp. 27-41. DOI:
[10.1007/s11123-020-00583-9](https://doi.org/10.1007%2Fs11123-020-00583-9).
URL:
[https://doi.org/10.1007/s11123-020-00583-9](https://doi.org/10.1007/s11123-020-00583-9).

<a name=bib-RADCLIFFE2021></a>[Radcliffe, B.](#cite-RADCLIFFE2021)
(2021). _Beeronomics: Factors affecting your print_. Investopedia.

<a name=bib-Salas2019></a>[Salas-Velasco, M.](#cite-Salas2019) (2019).
"Competitiveness and production efficiency across OECD countries". In:
_Competitiveness Review: An International Business Journal_ 29.2, pp.
160-180. DOI:
[10.1108/CR-07-2017-0043](https://doi.org/10.1108%2FCR-07-2017-0043).
URL:
[https://doi.org/10.1108/CR-07-2017-0043](https://doi.org/10.1108/CR-07-2017-0043).
---
<a name=bib-Rumankova2018></a>[Sanova, R. and P.
Kolar](#cite-Rumankova2018) (2018). _Beer Consumption in the Czech
Republic and its Determinants_. Czech University of Life Sciences
Prague, Faculty of Economics and Management.

<a name=bib-Swinnen2011></a>[Swinnen, J. F.](#cite-Swinnen2011) (2011).
_The Economics of Beer_. Oxford University Press. ISBN: 9780199693801.
DOI:
[10.1093/acprof:oso/9780199693801.001.0001](https://doi.org/10.1093%2Facprof%3Aoso%2F9780199693801.001.0001).
URL:
[https://academic.oup.com/book/1445](https://academic.oup.com/book/1445).

<a name=bib-Tremblay1985></a>[Tremblay, V.](#cite-Tremblay1985) (1985).
"A reappraisal of interpreting Rising Concentration: The case of Beer".
In: _The journal of Business_ 58.4, pp. 419 - 431.

<a name=bib-Wickham2016></a>[Wickham, H.](#cite-Wickham2016) (2016).
_ggplot2: Elegant Graphics for Data Analysis_. Springer. URL:
[https://ggplot2.tidyverse.org](https://ggplot2.tidyverse.org).

<a name=bib-Wickham2019></a>[Wickham, H., M. Averick, J. Bryan, et
al.](#cite-Wickham2019) (2019). _Welcome to the tidyverse_. , p. 1686.
URL: [10.21105/joss.01686.](10.21105/joss.01686.).

<a name=bib-Wickham2023></a>[Wickham, H. and J.
Bryan](#cite-Wickham2023) (2023). _readxl: Read Excel Files_. URL:
[https://github.com/tidyverse/readxl](https://github.com/tidyverse/readxl).

{\rtf1\ansi\ansicpg1252\cocoartf2513 \cocoatextscaling0\cocoaplatform0{\fonttbl\f0\fswiss\fcharset0 Helvetica;} {\colortbl;\red255\green255\blue255;\red199\green203\blue211;\red0\green0\blue0;\red255\green255\blue255; \red212\green20\blue102;\red20\green152\blue106;} {\*\expandedcolortbl;;\cssrgb\c81961\c83529\c85882;\cssrgb\c0\c0\c0;\cssrgb\c100000\c100000\c100000; \cssrgb\c87451\c18824\c47451;\cssrgb\c0\c65098\c49020;} \margl1440\margr1440\vieww10800\viewh8400\viewkind0 \deftab720 \pard\pardeftab720\partightenfactor0 \f0\fs28 \cf2 \cb3 \expnd0\expndtw0\kerning0

\ \ \

}