Public Goods Provision

# <rr> Public Goods Provision</rr>
## Economic Development & Construction 0008
### <img src="images/cover1.png" /><br> <a href="http://www.aniket.co.uk/">Dr. Kumar Aniket</a>
### Lecture 4

---

# Context

**Lecture 1**: Why economic growth is critical for improving living standards

**Lecture 2**: Mechanics of economic growth

**Lecture 3**: Complementarity between public goods and markets.

**Lecture 4**: Important of .embb[public goods provisions] in developing countries

???

This is just a lecture slide that recaps what we have done in the course till now. We started with the definition and examples of public goods in the last lecture. In this lecture we expand other notion of public goods.

---

# Refresher: Solow Growth Model

**Solow growth model**:

*Diminishing* .embb[marginal product of capital] limits economic growth potential

*Lower* the .embb[marginal product of capital], lower the investment, lower the economic growth of per-capita income (living standards)

Assumes **markets** (for consumer goods, capital goods and factor inputs) work .embb[perfectly], i.e., no transaction cost of accessing the market.

- Economy is **Pareto-efficient**, i.e., nothing economic surplus remains unexploited.

- Public goods that reduce the **transaction cost** of accessing the market can *potentially* lead to Pareto improvement 
	
---

# Refresher: Economic Growth and Markets

**Jensen, 2007**:

markets work better if transaction cost of accessing it is lower

- transaction cost for .embb[suppliers] *accessing the market*

- transaction cost for .embb[consumers] *accessing the market*

- Mobile phone networks increase fisherman's profits while reducing consumer's prices in Jensen (2007). .embb[Pareto improvment].

- A good road network can play a significant role in **reducing** the .embb[transaction cost of accessing markets] and encourage economic activity, i.e.,  **economic growth**.

---

Diminishing returns to capital constrains economic growth. Public goods have the potential of counteracting is diminishing returns to by encouraging economic activity.

We have already seen in Jenson 2007 one day in which this can be done through public goods. In Jensen 2007, mobile phones decreased the transaction cost of accessing the market for the fisherman. In doing so, they increase the profit margins for fishermen as well as decrease consumer prices.]

---

# Refresher: Public Goods

- public goods funded through **tax resources** and **free** to use

- no one would pay for goods that are non-excludable
	
	- the marginal cost of producing a non-rival good is zero

---

# Refresher: Public Goods

.thebox[
.embb[Public goods] are critical in **lowering** the .emab[transaction cost of accessing the market]]

-  e.g. road, canals, mobile phone networks, transport infrastructure

- *health* and *education* makes **labour** more productive

- law and order, electricity, banks and post-offices makes **capital** more productive

---

It is important to understand what is non-rival and non-excludable goods. Read the Core Econ. Section 4.6. and Section 12.5 for examples of public goods.

From the perspective of canonical models, public goods increase the productivity of factor inputs. For example, firms will have no access to markets without roads. The better the road network, the more cheaper it is for firms to access the markets. Cheaper access to markets makes the firm more profitable, which in turn, encourages capital investment want.

A similar story can be told about how public good like schools, universities and hospital the decrease the cost of acquiring human capital and in doing so in increase the productivity of labour.]

---

# Urban-rural Divide in Public Goods

- cheaper to provide public goods in areas where **population density** is high

- urban areas have more **economic activity**, pay more taxes and hence are vocal in demanding more public goods

- rural areas cannot **develop economically** due to lack public goods like roads, banks, post-offices, electricity, schools and hospitals

---

It is cheaper to provide public goods to urban areas because of a higher population density. In rural areas, there public good has to coverage large geographical span with very low usage.

For instance, government can provide street lighting for a 1 million people in an urban area at a fraction of the cost it would take to provide street lighting to 1 million people living in the rural area.]

---

---

# Government's production function

**Inputs:** *public capital* and *government employees*. Government pays for the inputs through tax revenues

**Outputs:** Public goods and services that are free to use<sup>1</sup>

&nbsp;

.pull-left[.ema[**Public capital** is the capital owned by government. Since government represents the people, it is in principle collectively owned by the public]]

.pull-right[.ema[**Public goods** & are the range of goods and services produced by the government using public capital and government employees that people living in the country are free to use]]

.footnote[
[1] Given that public goods and services are non-excludable, it would be impossible the government to extract a direct price for it. Goods have to be made at excludable in order for government to charge for it.]

---

Just the way firms produce goods using the capital and labour, government produces public goods using public capital and government employees (labour employed directly by government). Government pays for both accumulating public capital and hiring government employees through the tax revenues it obtains from taxing individuals and firms in the economy.]

---

# Public goods versus markets

Should the government provide public goods or leave the provision to the market?

**Pros:**

- Market fails to provide non-rivalry and non-excludable goods
- Adequate provision of public goods creates well-functioning markets, makes capital and labour more productive, which encourages per-capita economic growth

- **Cons:** without an incentive to reduce cost, the government is inefficient at providing public goods.

---

# Public goods Examples

- .emab[Examples] of publicly provides goods and services

- Road building in Kenya. **Burgess et al. (2015)**.
	
	- Regulatory framework for mobile network. **Jensen (2007)** shows how public good provision and private capital can potentially be .embb[complementary].
	
	- Schools and hospitals to the .embb[poorest of the poor]. Schools and hospitals in .embb[rural areas]. **Chaudhury et. al (2016)**
	
	- Preventing deforestation
	
	- National defence
	
	- Law and order
	
	- Constitution

---

Public goods facilitate economic activity on one hand, yet, government is inefficient because it does not have a clear motive the way firms have profit motive. The political leadership may have an incentive to get reelected in democracy, yet large number of government employees have nothing to gain from the political leadership getting reelected.

Government pays everyone similar salaries and there is very little pecuniary motive to excel in government. Government largely depends on its employees motivation.]

---

---

# Kenya's Political System

Kenya got independent in 1963 and moved back and forth between democratic and autocratic regimes

Since 1993, the democratic norms are reflected in greater political freedom and freedom of press

&nbsp;

Year | President | President's Tribe | Political System
:---|:---|:---|:---
1963-1969 | Kenyatta | Kikuyu | Democracy
1970-1978 | Kenyatta | Kikuyu | Autocracy
1979-1992 | Moi | Kalenjin | Autocracy
1993-2002 | Moi | Kalenjin | Democracy
2003-2011 | Kibaki | Kikuyu | Democracy

???

The main thing to notice is that Kenya was democracy before 1970 and after 1993. It was a autocracy between 1970 and 1993.

---

# Economic Growth in Kenya

Kenya's per-capita economic growth was higher than rest of Sub-Saharan Africa during 1st phase of democracy. Autocracy's had a catastrophic long-term impact on economic growth.

---

**Autocracy**: a system of government by one person with absolute power.

Policy enacted in a particular period affects the economic activity in a country with a lag of about 5 to 10 years. The previous figure shows the impact autocracy (from 1970 to 1992) had on lower Kenya's growth potential.

Kenya became autocratic in 1970 when it was growing at a much faster rate than the rest of the Sub-Saharan African countries. Autocracy lowered the economic growth rate in Kenya and by the end of the autocracy in 1992, Kenya's economic growth rate was close to the rest of the Sub-Saharan Countries. Kenya has struggled to shake of problems of its autocratic era and regain its economic growth rate from early 1970s.

In the next few slide, we will see how autocracy subverted the provision of public goods in Kenya and the negative impact it has on economic activity.]

???

---

# Autocracy and Growth

Why did Kenya under-perform its neighbours during .embb[autocracy]?

- the leaders have .emab[discretion] over **government expenditure** during autocracy

- leads to .emab[sub-optimal] **public goods provisions**, i.e., it does not respond to the socio-economic needs of the population

- .embb[sub-optimal allocation of public goods], e.g. roads

- **high transaction costs** of accessing markets lead to dysfunctional markets

- lowers **marginal productivity of private capital**, i.e., entrepreneurs don't invest
	
	- lowers **economic growth**
	
???

While democracy has various checks and balances on the power of the leaders. In democracy, the leaders have incentive to provide public goods where it is needs. Their discretion is limited discretion. Autocracy gives the autocratic leader much greater discretion in terms of where and how they provide public goods. The leaders can reward one part of the society at their discretion and deprive other parts of the society.

In Singapore, Lee Kuan Yew, its autocratic leader provided public goods in such a way that it encouraged economic activity across the society and gave everyone an opportunity to enrich themselves equally. As we will see in the next few slides, this was not the case in Kenya. In Kenya, its leaders directed the government to provide public goods provision for their own ethic tribes and deprive the other ethnic tribes.]

---

# Ethnic favouritism in Africa

.embb[Ethnic favouritism]: when members of a particular ethnic group control the government and influence government expenditure in order to disproportionately benefit members of their own group.

**Collier and Gunning (1999)**: One of the reasons for Africa’s lack of economic progress is its pattern of .embb[ethnic favouritism]

---

While ethnic favouritism has often been blamed for lack of Africa's economic progress, finding clear evidence for it has been difficult. Burgess et al. (2015) find evidence of exactly this in terms of how road building was carried out in Kenya during the periods of autocracy and democracy.]

---

# Measuring Ethnic Favouritism

- If government provides better public goods in area A and not in area B, people would just **move** to area B.

- This would change the ethnic composition of the area making ethnic favouritism **empirically invisible**.

- To find empirical evidence, we need a **stable** population that has not responded to uneven provision of public goods for long periods of time

???

As we will see, road building was done fairly in periods of democracy and done in way to favour the ethnic tribes linked to the autocratic leader during autocracy.

---

# President's Road

.emai[Finding empirical evidence] requires finding a pattern of population that is segregated and has remained **stable** over time

**Burgess et al. (2015)** examine whether road building in Kenya favoured certain ethnicities over others

- Kenya's population is segregated along tribal lines and population composition has not changed over time

- Stark ethnic segregation in Kenya due to its colonial legacy.

- Each district dominated by a single ethnic group

- Pattern has remained stable over time: people have not moved in response to the pattern of government expenditure 
	
---

# Research Question

Allows Burgess et al. (2015) to ask the following question:

1. Does the President influence the government to build more roads in districts where people of his ethnicities live?

2. Does the nature of the political system influence road building?

]

---

Kenya has a very settled ethnic pattern and this is due to its colonial legacy where people were encouraged/forced to stay in the areas where their own ethnic tribes were dominant. It was very difficult or impossible for people to move to other areas. Hence, Kenya was and still remains largely ethnically segregated and its easy to identify the dominant ethnic groups in each area.

This allows Burgess et. al. (2005) to examine whether the President's build roads in the areas where their own ethnic group lives and whether this patters varies across democratic and ethnic eras. ]

---

# Road Population Share Ratio

`$\gamma_i$` measures whether road expenditure in district reflects it population share

`$$\gamma_i=\frac{\text{proportion of road expenditure spent in district}~i}{\text{proportion of total population living in district}~i}$$`

`$\gamma_i$` | Treatment of the district for roads
---|---
`$\gamma_i>1$` |  district getting favourable treatment 
`$\gamma_i=1$` | district getting fair treatment
`$\gamma_i<1$` | district getting unfavourable treatment

&nbsp;

.pull-left[
*President's co-ethnic districts*: districts that share their ethnicity with the President]

.pull-right[
**Measuring ethnic favouritism**: plotting average `$\gamma_i$` for President's co-ethnic districts]

---

Road population share ratio (RPS ratio) is obtained by dividing the share of *total road expenditure spent in the district* with its *population share.*

For example, let's say 2% of the Kenya's population living in district A. If the 3% of the total road expenditure went to district A, then the Road Population Share Ratio is 1.5, which implies the district is getting favourable treatment as compared to the rest of the districts. If 1% of the road expenditure is spent in that district, it means that RPS ratio is 0.5 and its getting less favourable treatment. District A is deemed to get even treatment if it gets 2% of the road expenditure so that its RPS ratio is 1. ]

---

# Measuring ethnic favouritism

---

Ratio of district share of road expenditure to population share in districts that share their ethnicity with the President is a measure of ethnic favouritism.

As we can see, during autocracy, the road population share ratio is much greater than 1 for co-ethnic districts, i.e., districts that share their ethnicity with the President and less than 1 for non co-ethnic districts, i.e., districts that don't share their ethnicity with the president.

The pattern by and large is that during democracy the road population share ratio of co-ethnic and non co-ethnic districts were quite close to each other (with some aberrations at the margin).

]

---

# Measuring ethnic favouritism

Plotting average `$\gamma_i$` for **Kikuyu**, **Kalenjin** and **other districts**

---

This graph in the slides presents a more detailed picture of ethnic favouritism and finds that certain tribal areas benefit quite significantly if the President shares their ethnicities with them. The change in ethnic favouritism with change in Presidency in 1979 is particularly remarkable. ]

---

# Main results of Burgess et. al (2006)

*Road building* in Kenya subject to a high degree of .embb[ethnic favouritism] during *autocracy*

- President usually directed road spending to his co-ethnic districts

- These districts usually received three times the average road expenditure and five times the length of roads during periods of autocracy.

- Groups not linked to the president constitute the bulk of the population and receive far fewer roads during autocracy

.embb[Ethnic favouritism] **disappears** during *democracy* and road building benefits all districts evenly

- This reversal is not enough to overturn the **road deficit** accumulated during the period of autocracy

---

The main point of the paper is that *ethnic favouritism impacts public goods provision*. Areas that are out of favour at some period **accumulate a road deficit** and find it difficult to reverse that deficit over time.

Roads are critical for markets to function properly. Road deficit leads to areas lagging behind in terms of economic activity as compared to other parts of the country.]

---

---

# Evolution of Per-capita output across Africa

---

There is a huge variation in the way per-capita income has evolved across Africa. Understanding the reasons for varying economic growth experience is an active area of research in economic development. ]

---

# Economic Growth across Africa

&nbsp;

Variety of growth experiences

Sharp increase in per-capita output since 2000

Cote d'Ivoire, Kenya, Ethopia and Uganda are African success stories

Democratic republic of Congo and Seirra  Leone are exceptions]

---

The economic growth experience is directly linked with investment patterns, which as we know from Solow Growth model and Aniket (2018) is related to the both the *saving investment channel* that facilitates the private capital accumulation and the *fiscal channel* that facilitates the public goods provisions. ]

---

# Collier (1999)

- **Investment as proportion of GDP** is 18% in Africa<sup>1</sup>

- Investment as proportion of GDP is 23% in South Asia and 29% in lower middle income countries

- Africa has a unique geography that is a disadvantage in realising its growth potential

- **low population density**
	
	- **Mountainous terrain**
	
	- **landlocked** areas with few major waterways

.footnote[
[1] Remember Investment as proportion of GDP measure the proportion of output that is channeled towards capital accumulation. High investment ratio is associated with high per-capita output.]

---

Collier (1999) stresses that there is clear linked between the low investment rate in Africa as compared to South Asia or the lower middle income country.

The investment ratio is simply the proportion of GDP that is saved in banks and then lent out to the entrepreneurs in economy that have profitable investment projects. If 29% of the total output becomes investment in middle income countries and only 18% in the African countries, then African countries are accumulating far less capital per-capita and hence missing out on the opportunities to grow fast economic growth. 
]

---

# Raising Population density,

One solution is to raise population density through large-scale urbanisation

Increasing population density has the potential to radically reduce per capita cost of public goods (infrastructure) provisions

Policies that could raise density.

- Prioritising **transport infrastructure** for urban connectivity

- Increasing **residential density** within cities by improving the operation of housing and urban land markets

---

The story is not simply that there is not enough investment. It is more complicated. The story is that the geographical features in Africa make investment more difficult across space that is inhospitable for economic activity. The only way in which these large public spaces can be tamed is through public goods, yet, there is not enough investment in public goods.

Recall how mobile phone network as a public good allowed fisherman to counter-act the problems imposed by space. Transport and communication networks allow people to conduct economic activity across large spaces. ]

---

# Cost of Urban Public Goods

---

Public good require tax revenues. Developing countries by their definition lack tax revenues to create public good. Hence, it is logical that it start with the most cost-effective public goods, i.e., public goods that give them the most bang for the money in terms of its impact on economic growth of the country.

The table in the previous slides is some Collier and Venables (2016)<sup>1</sup>.

]

.footnote[
[1] P. Collier and A. J. Venables. Urban infrastructure for development. Oxford Review of Economic Policy, 32(3):391–409, 2016.
]

---

# Collier and Venables (2016)

Cities are potentially high-productivity areas and drivers of economic growth.

> **spatial configuration of economic activity** in the city is dependent on **infrastructure**, as the investment decisions of firms and households respond to the transport network.

> urban connectivity creates an environment where firms and workers can become specialised and can reach an **efficient scale of operation**.

---

The solution is to start with urbanisation. It is easier to provide public goods and facilitate economic activity through creation of market in urban areas. Urban agglomerations are always the first step in creation of markets for a developing countries. Once the economic growth picks up and the country has enough resources, it can start providing public goods to the rural areas.]

---

# Collier and Venables (2016)

Poor infrastructure raises the cost of doing business in a city.

.pull-left[
Many cities in East Asia and South Asia—have been able to attract and grow the sectors in .emab[high density areas] of economic activity]

.pull-right[
In contrast, most African cities are .emab[artisanal cities], characterised by *low levels of investment in infrastructure* and *low levels of residential and business investment*]

.footnote[ Largest cities in the world [ම](https://www.worldometers.info/population/largest-cities-in-the-world/) ]

---

East Asia and South Asia has fruitfully used the urbanisation as a strategy to kick start economic growth. In contrast, African has not benefited from agglomeration. For Africa, the one way to kick start growth would be concentrate on providing public goods in urban areas where its is most cost-effective. Burgess et. al. (2015) show how the patterns of ethnic favouritism is a hinderance. It allocates public goods for reasons other that ones that would enrich the whole society in the long-run. ]

---

# Schools and Hospitals

They also determine the **productivity** of workers and thus are critical inputs in economy's production function.

- while the middle-class may afford to pay for privately provided health and education services, the **poorest of the poor** entirely depend on the public provision of these services

- Public provision of these services is cheaper in urban areas and particularly challenging in remote **rural** areas

---

# Addressing Absence

> "With one in five government primary-school teachers and more than a third of health workers **absent** from their facilities, developing countries are wasting considerable resources and missing opportunities to **educate** their children and improve the **health** of their populations." &nbsp;&nbsp;&nbsp;-- Chaudhury  et. al. (2016)

.emdb[Chaudhury  et. al. (2016)] document how it more difficult to both health and education public goods in areas that are difficult to access.

- **Transport links** are critical in reducing the transaction cost of delivery of health and education services

.embb[Public provision] of health and education services requires funding through taxes (Aniket, 2018)

- Areas with economic activity **subsidise** areas where there is not economic activity. The tax base severely limits the government's ability to fund these services.

---

# Conclusion

**Private investment** is critical for private capital accumulation

Private investment requires **public goods provision**: lack of adequate public goods provision (infrastructure) can stop a country from realising its growth potential

- reducing transaction cost of market access is critical

- public goods that increase the productivity of labour and capital are critical

Africa's **low population density** increases the cost of adequate public goods provisions

**Autocracy** creates further imbalances in the way public goods are provided

Effective **urbanisation** accompanied by high quality infrastructure provision is one of the solutions for Africa

---

The main emphasis in the lecture is the role public goods play in facilitating accumulation of private capital. As we know from Solow Growth model, accumulation of public goods is critical for increasing per-capita income.

Africa has certain geographical features that make it inhospitable for economic activity. This implies that returns to capital are lower and may also diminish at a faster rate than in other continents.

If market places are scarce, it implies that the transaction cost of accessing the market is high for both buyers and sellers. It then directly implies that the returns to capital i.e., marginal product of capital is lower.]

---

There are developed countries like Norway, Canada and Australia which also have geographical features that are inhospitable for economic activity. These countries have been able to counter-act this inhospitable geography and the diminishing returns to capital through public goods, i.e., a extensive transport and communication network that allows markets to work efficiently and thus counter-acts the diminishing returns to capital. If Africa is to turn the corner, it is critical that it finds some way to counter-act its inhospitable geography.

Public goods provision is one solution. Given that public goods are cheaper to provide when population density is high, urbanisation is one of the solutions for Africa. Urbanisation requires creating areas that are well stocked with public goods and centres of intense economic activity. South Asia and East Asia in the recent past and Europe before that has successfully used this strategy to increase their per-capita income. There is no reason why Africa can't replicate the same process. ]

<style>

h1, h2, h3 {
  color: #EC5800; 
}

p {
	line-height: 1.5em;
}

rr {
	color: #002E63;
}

a {
  color: #002E63; 
}

.inverse {
  background-color: #D2691E; 
}

.tab { 
       display:inline-block; 
       margin-left: 15px; 
}

.ema  {color: rgb(43,106,108);}
.emb  {color: rgb(184,13,72); }
.emc  {color: #1034A6;        }
.emd  {color: rgb(64,64,64);  }
.eme  {color: #614051;  }
.emg  {color: #6D351A;  }
.emgr  {color: #696969;  }
.emn  {color: #796878;  }

.emo  {color: rgb(229,65,6);  }
.emob {color: rgb(229,65,6); font-weight: bold;}
.emoi {color: rgb(229,65,6); font-style: italic;}

.emz  {color: #004225;  }
.emzb {color: #004225; font-weight: bold;}
.emzi {color: #004225; font-style: italic;}

.emab  {color: rgb(43,106,108);font-weight: bold;}
.embb  {color: rgb(184,13,72); font-weight: bold;}
.emcb  {color: #1034A6;        font-weight: bold;}
.emdb  {color: rgb(64,64,64);  font-weight: bold;}

.emai  {color: rgb(43,106,108);font-style: italic;}
.embi  {color: rgb(184,13,72); font-style: italic;}
.emci  {color: #1034A6;        font-style: italic;}
.emdi  {color: rgb(64,64,64);  font-style: italic;}

.emdb  {color: rgb(64,64,64);  font-weight: bold;}

.footnote {color: gray;}

.red { color: red; }

.mybox  {
	color:#3D2B1F;
	background-color:#3D2B1F10;
	margin:1em;
	padding: 1em;
	border-radius: 10px;
}

.thebox  {
	color:#704214;
	background-color: #F5DEB320; 
	margin: 0.1em 0.5em 0.1em 0.5em;
	padding: 0.1em 1em 0.1em 1em;
	border-radius: 10px;
	border-color: #704214; 
	border-style: solid; 
	border-width: 2px; 
}

</style>