Top Banner
INTRODUCTION ELASTICITY: Elasticity refers to the relative responsiveness of a supply or demand curve in relation to price: the more elastic a curve, the more quantity will change with changes in price. In contrast, the more inelastic a curve, the harder it will be to change quantity consumed, even with large changes in price. For the most part, Goods with elastic demand tend to be goods which aren't very important to consumers, or goods for which consumers can find easy substitutes. Goods with inelastic demands tend to be necessities, or goods for which consumers cannot immediately alter their consumption patterns. Elasticity of demand measures the responsiveness of change in quantity demanded of a good because of change in prices. If a curve is more elastic, then small changes in price will cause large changes in quantity consumed. The concept of elasticity is used to mathematically and thus, precisely measure the changes in demand in quantitative terms. The demand depends upon various factors such as the price of a commodity, the money income of consumer the prices of related good the taste of the people etc.The elasticity of Demand measures responsiveness of quantity demanded to a change in any one of the above factors by keeping other factors constant. The term elasticity of demand measures the extent of changes in the quantity demanded of a given commodity as a result of the change in the demand in the determinant. If that is so, the types 1 | Page
31
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: ECO REPORT.docx

INTRODUCTION

ELASTICITY:

Elasticity refers to the relative responsiveness of a supply or demand curve in relation to price: the

more elastic a curve, the more quantity will change with changes in price. In contrast, the more

inelastic a curve, the harder it will be to change quantity consumed, even with large changes in

price. For the most part, Goods with elastic demand tend to be goods which aren't very important to

consumers, or goods for which consumers can find easy substitutes. Goods with inelastic demands

tend to be necessities, or goods for which consumers cannot immediately alter their consumption

patterns. Elasticity of demand measures the responsiveness of change in quantity demanded of a

good because of change in prices. If a curve is more elastic, then small changes in price will cause

large changes in quantity consumed.

The concept of elasticity is used to mathematically and thus, precisely measure the changes in

demand in quantitative terms. The demand depends upon various factors such as the price of a

commodity, the money income of consumer the prices of related good the taste of the people

etc.The elasticity of Demand measures responsiveness of quantity demanded to a change in any

one of the above factors by keeping other factors constant.

The term elasticity of demand measures the extent of changes in the quantity demanded of a given

commodity as a result of the change in the demand in the determinant. If that is so, the types of

demand elasticities must equal the number of demand determinants. However, from the managerial

and business point of view, only few of the demand determinants are given serious consideration

and therefore the elasticities that we undertake to study are:

TYPES OF ELASTICITY OF DEMAND

Price Elasticity of demand.

Income Elasticity of demand.

Cross Elasticity of demand.

Advertisement Elasticity of demand.

1 | P a g e

Page 2: ECO REPORT.docx

DETERMINANTS OF ELASTICITY

1. NATURE OF THE COMMODITY :

Human’s wants, i.e. the commodities satisfying them can be classified broadly into

necessaries on the one hand and comforts and luxuries on the other hand. The nature of

demand for a commodity depends upon this classification. The demand for necessities is

inelastic and for comforts and luxuries it is elastic.

2. NUMBER OF SUBSTITUTES AVAILABLE :

The availability of substitutes is a major determinant of the elasticity of demand. The large

the number of substitutes, the higher is the elastic. It means if a commodity has many

substitutes, the demand will be elastic. As against this in the absence of substitutes, the

demand becomes relatively inelastic because the consumers have no other alternative but to

buy the same product irrespective of whether the price rises or falls.

3. NUMBER OF USES :

If a commodity can be put to a variety of uses, the demand will be more elastic. When the

price of such commodity rises, its consumption will be restricted only to more important

uses and when the price falls the consumption may be extended to less urgent uses, e.g. coal

electricity, water etc.

4. POSSIBILITY OF POSTPONEMENT OF CONSUMPTION :

This factor also greatly influences the nature of demand for a commodity. If the

consumption of a commodity can be postponed, the demand will be elastic.

5. RANGE OF PRICES :

The demand for very low-priced as well as very high-price commodity is generally

inelastic. When the price is very high, the commodity is consumed only by the rich people.

A rise or fall in the price will not have significant effect in the demand. Similarly, when the 2 | P a g e

Page 3: ECO REPORT.docx

price is so low that the commodity can be brought by all those who wish to buy, a change,

i.e., a rise or fall in the price, will hardly have any effect on the demand.

6. PROPORTION OF INCOME SPENT :

Income of the consumer significantly influences the nature of demand. If only a small

fraction of income is being spent on a particular commodity, say newspaper, the demand

will tend to be inelastic.

From the above analysis of the determinants of elasticity of demand, it is clear that no precise

conclusion about the nature of demand for any specific commodity can be drawn. It depends upon

the range of price, and the psychology of the consumers. The conclusion regarding the nature of

demand should, therefore be restricted to small changes in prices during short period. By doing so,

the influence of changes in habits, tastes, likes customs etc., can be ignored.

INCOME ELASTICITY

Income Elasticity is a measure of responsiveness of potential buyers to change in income. It shows

how the quantity demanded will change when the income of the purchaser changes, the price of the

commodity remaining the same.

It is equal to unity or one when the proportion of income spent on good remains the same even

though income has increased.

It is said to be greater than unity when the proportion of income spent on a good increases as

income increases.

It is said to be less than unity when the proportion of income spent on a good decreases as income

increases.

Generally speaking, when our income increases, we desire to purchase more of the things than

we were previously purchasing unless the commodity happens to be an “inferior” good.

Normally, then, since the income effect is positive, income elasticity of demand is also positive. It

is zero income elasticity of demand when change in income makes no change in our

purchases, and it is negative when with an increase in income, the consumer purchases less, e.g., in

the case of inferior goods.

3 | P a g e

Page 4: ECO REPORT.docx

It may be carefully noted that for any individual seller or firm, the demand for the product as a

whole may be inelastic. By lowering the price, as compared with his rivals, the seller can

infinitely increase the demand for his product. The demand curve will thus be a horizontal line.

Elasticity, viz., price elasticity and income elasticity, are valuable aids in the measurement of

demand for different commodities. As such they are also helpful in measuring the incidence of

taxation.

In economics, income elasticity of demand measures the responsiveness of the demand for a good

to a change in the income of the people demanding the good, holding all prices constant. It is

calculated as the ratio of the percentage change in demand to the percentage change in income.

Income elasticity of demand can be used as an indicator of industry health, future consumption

patterns and as a guide to firms investment decisions.

The income elasticity of demand can be classified in to five types on the basis of values of its co-

efficient. These are as follows:

1. Negative income elasticity of demand(ey <0)

2. Zero income elasticity of demand (ey =0)

3. Income elasticity of demand less then unity (ey <1)

4. Income elasticity of demand equal to unity (ey =1)

5. Income elasticity of demand greater then unity (ey >)

Now little explanation regarding the above:

1. INCOME ELASTICITY OF DEMAND GREATER THAN ONE:

When the percentage change in demand is greater than the percentage change in income, a

greater portion of income is being spent on a commodity with an increase in income-

income elasticity is said to be greater than one.

4 | P a g e

Page 5: ECO REPORT.docx

2. INCOME ELASTICITY IS UNITARY:

When the proportion of income spent on a commodity remains the same or when the

percentage change in income is equal to the percentage change in demand, EY = 1 or the

income elasticity is unitary.

3. INCOME ELASTICITY LESS THAN ONE (EY< 1):

This occurs when the percentage change in demand is less than the percentage change in

income.

4. ZERO INCOME ELASTICITY OF DEMNAD (EY=o):

This is the case when change in income of the consumer does not bring about any change

in the demand for a commodity.

5. NEGATIVE INCOME ELASTICITY OF DEMAND (EY< o):

It is well known that income effect for most of the commodities is positive. But in case of

inferior goods, the income effect beyond a certain level of income becomes negative. This

implies that as the income increases the consumer, instead of buying more of a commodity,

buys less and switches on to a superior commodity. The income elasticity of demand in

such cases will be negative.

THE FORMULA FOR CALCULATING INCOME ELASTICITY:

5 | P a g e

Page 6: ECO REPORT.docx

6 | P a g e

Page 7: ECO REPORT.docx

LITERATURE REVIEW

While the popular view is that public transit is an inferior good, those studies that have included

income in their studies of the behavior of consumers of public transit have produced results that are

either ambivalent or do not confirm to this view. Nonetheless, no study has yet tried to address this

specific problem as the main focus.

Various authors have found widely diverging results for the income elasticity of demand for busing

routes, as detailed by Holmgren (2007). He found using a meta-analysis that estimates on the

income elasticity of demand for public buses were ambiguous and highly dependent on the demand

specifications included. He found that while some studies had found negative income elasticities of

demand, some had also found positive results, and that the overall average was 0.17.

Several authors have investigated the demand for public transportation on a route by route

basis.Schmenner (1975) pursued a methodology of restricting his area of study to populations

within two city blocks of bus routes in three Connecticut cities. The study found that in bus transit

with the log of revenue per mile or revenue per hour as the dependent variable, the sign on the log

of family income was positive when all three cities in his study were pooled. When performed on a

city by city basis, the sign on family income fluctuated, seeming to indicate that city-specific

factors were key drivers behind this result. Further, he claimed that previous studies had suggested

that demand for busing was price inelastic.

Schmenner’s finding is important because it suggests that if a relatively slow form of public transit

would in fact have a positive income elasticity of demand, than faster forms would also. Glaeser,

Kahn, and Rappaport (2006) found that the fixed time-cost of subways is less than that for bus

transit, and that subways had on the whole a “much lower” time-cost per mile. The paper also

found that when surveying all modes of public transit with 2000 census tract data in Boston,

Chicago, New York, and Philadelphia that there was a positive correlation between the log of

income and public transit usage for fixed distances outside of the Central Business District. When

changing the urban mix to Houston, Atlanta, Pheonix, and Los Angeles, the authors found that the

7 | P a g e

Page 8: ECO REPORT.docx

correlation was negative. Differing levels of urban residential and employment concentrations

seem to produce different patterns of transit usage according to this study. While the first set of

cities was specifically selected to include subway transit, the study did not attempt to find separate

results on income for rail and bus transit.

A factor that could influence this outcome is discussed by Glaeser, Kahn, and Rappoport (2006) is

the age of the cities and their transit networks as an important factor in the relationship between

income and public transit usage. Baum-Snow, Kahn, and Voith (2005) found that in networks that

have been built or extended between 1970 and 2004, poorer census tracts were 20.6% more likely

to have gained access to rail. Later day rail transit operations and expansions are funded and

directed by local, state, and federal governments.The greater likelihood for poorer tracts to receive

new rail transit could be explained as a purposeful decision by policy makers. This is in contrast to

the first period of rail transit construction, which was largely initiated by private companies. Later-

day urban rail transit expansions are sometimes constructed as a policy tool to address, in part,

problems of urban poverty. For example, Gilderbloom and Rosentraub (1990) specifically

recommend that policy makers utilize mass transit as a method for improving opportunities and

promoting independence of the poor, elderly, and the disabled in the Houston area.

Another cause that could drive this finding is that the “newer” cities are much less centralized and

dense than older ones. In a paper by Anas, Arnott, and Small (1998), they explain that in cities who

saw most of their growth prior to the invention of the automobile, the rich outbid the poor for the

most centrally located living spaces. With the construction of electric streetcars, the rich moved

outwards from the city center and settled along mass transit line to create the first “streetcar

suburbs”. Cities whose growth was driven by the construction of radial freeways or at the

beginning of the streetcar period are far more dispersed than those settled before. For example,

Poulton (1980) points out that Los Angeles developed the world’s largest streetcar system which

allowed for greater dispersal, a facet also remarked upon by others, such as Gordon and Richardson

(1996). There exists some evidence that cities that construct subway lines can spur increased

residential densities along the mass transit corridors, even in cities where automobile use is

widespread, such as in Davies (1976). Yet even in “newer” cities, there is some evidence that the

role of income as a predictor of public transit usage cannot be assumed to automatically reverse.

Dajani, Egan, and McElroy (1975), conducted a study using the Metropolitan Atlanta Regional

Transit Authority’s transit planning studies showing that when dividing up the metro Atlanta area

into zones, the coefficient on family income was positive, but not statistically significant. The key

driver of rail transit usage seemed to be distance to the nearest transit station which was statistically

significant at the 0.01 level. The chief drawback of this study was that it measured net benefits

from the presence of a heavy rail system as opposed to ridership directly, and had few degrees of

8 | P a g e

Page 9: ECO REPORT.docx

freedom. Nonetheless, it is somewhat counterintuitive that Atlanta, which has relatively lower costs

of daily parking (an indicator of employment concentration in the CBD), would show a positive

relationship between net benefits and income. On a city-wide basis, further evidence is

inconclusive about the sign on the income elasticity of demand for transit. Schenker and Wilson

(1967) showed that when compared across twenty-three metropolitan areas, the sign on family

income was positive, but not statistically significant. However, also using 1960 Census data,

Meyer, Kain, and Wohl (1965) found that the sign on income on public transportation usage was

negative when doing a cross-sectional comparison that did not involve econometric analysis.

The theoretical basis of all of these studies is centered on the premise that a city is monocentric in

nature. This assumes that all residents of a city commute to the central business district for their

employment. There are obvious problems with this assumption. Glaeser and Kahn (2004) and

Anas, Arnott, and Small (1998) have found that only 75.9% of metropolitan area employment in

the year 2000 was within three miles of the CBD. Available evidence indicates, however, that mass

transit commuters tend to overwhelmingly work in the CBD.

Rothenberg Pack (1992) states that 70% of SEPTA commuters work in downtown Philadelphia.

Baum-Snow, Kahn, and Voith (2004) also indicated that the vast majority of public transit users

were commuting to the CBD across the sixteen cities under study using data from the 1990 Census.

A casual study of the layout of these systems reveals why. Most are designed as a spoke and wheel

system, where the lines radiate outwards from the Central Business District. With the exception of

New York’s G Train and Philadelphia’s Norristown High Speed Line, all heavy rail lines in the

United States run towards or very near the Central Business District. While this is somewhat less

true of light rail lines, the easy majority also follow the same pattern. For this reason, like other

authors incorporating the monocentric city model into studies of public transit, the model is a

reasonable approximation of how rail transit commuters behave.

9 | P a g e

Page 10: ECO REPORT.docx

METHODOLOGY:

RESEARCH METHOD:

The research conducted descriptive research in which the focus is on the description of the

variables in the problem.

DATA COLLECTION:

Primary Data:

The Major source of primary data collection is through questionnaire.

Secondary Data:

The secondary data is collected through internet and articles

MEASUREMENT TECHNIQUE:

This research is based on 20 questionnaires for the people who use public transports, which contain

12 questions and the resulted data was then tested against the assumptions made.

RESPONSE FORMAT:

The questionnaires contain 12 questions, all of which were Closed-Ended questions.

SAMPLE PROCESS AND SAMPLE UNIT:

The sampling unit is based on those people who use public transport

10 | P a g e

Page 11: ECO REPORT.docx

QUOTA SAMPLING

Questionnaires of Consumers:

GENDER RESPONDENT

Male 10

Female 10

Total 20

Respondent

MaleFemale

11 | P a g e

Page 12: ECO REPORT.docx

QUESTIONNAIRE

Hello, Please help us in completing our Economic Analysis Project by giving

your Valuable time in filling this questionnaire

PART A (DEMOGRAPHICS)

1) Name__________________________________

2) Your Gender

Male

Female

3) What is your age?

18-25

26-30

30 +

4) Your Occupation?

Student

Own Business

Employer

5) Your income is

1000-5000

6000-10000

10000+

12 | P a g e

Page 13: ECO REPORT.docx

PART B (INCOME ELASTICITY)

1) Do you have Private transport?

Yes

No

2). How do you mostly travel?

Private Transport

Public Transport

3) How often do you use public transportation?

Daily

Sometime

4) Is public Transport cheaper than Private transport?

Yes

No

5) Do you feel save while travelling through Public transport?

Yes

No

6) If you get 20% increase in your salary/pocket money you will travel by?

Public Transport

Private Transport

7) If you get 20% decrease in your salary/pocket money you will travel by?

Public Transport

Private Transport

13 | P a g e

Page 14: ECO REPORT.docx

FINDING AND ANALYSIS

Q: 1: Age of the respondent

Aim: The Aim of this question is to know that from which age group questionnaire has been filled

Response Format: It is a Multiple choice question.

No of respondent to answer this question: 20

No of answers generated: 20

TABULATED RESULT

14 | P a g e

AGE Male Female

18-25 9 10

26-30 1 0

30+ 0 0

TOTAL 10 10

Page 15: ECO REPORT.docx

GRAPH

18-25 26-30 30+0

2

4

6

8

10

12

10

0 0

Chart Title

MaleFemale

15 | P a g e

Page 16: ECO REPORT.docx

Q: 2: Income of respondent?

Aim: The Aim of this question is to know that what the income of the respondents is.

Response Format: It is a Multiple choice question.

No of respondent to answer this question: 20

No of answers generated: 20

TABULATED RESULT

GRAPH

1000-5000 6000-10000 10000+0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

MaleFemale

16 | P a g e

INCOME Male Female

1000-5000 3 5

6000-10000 5 3

10000+ 2 2

TOTAL 10 10

Page 17: ECO REPORT.docx

Q: 3: How respondent mostly travel?

Aim: The Aim of this question is to know that how respondent mostly travel

Response Format: It is a multiple choice question.

No of respondent to answer this question: 20

No of answers generated: 20

TABULATED RESULT

MODE OF

TRANSPORTATION

MALE FEMALE

PRIVATE TRANSPORT 5 6

PUBLIC TRANSPORT 5 4

TOTAL 10 10

GRAPH

PRIVATE TRANSPORT PUBLIC TRANSPORT0

1

2

3

4

5

6

MALE FEMALE

17 | P a g e

Page 18: ECO REPORT.docx

Q:4 If you get 20% increase in your Salary/ Pocket money you will travel by?

Aim: The Aim of this question is to know choice of travelling with 20% increase in Salary/Pocket

Money.

Response Format: It is a multiple choice question.

No of respondent to answer this question: 20

No of answers generated: 20

TABULATED RESULT

MODE OF

TRAVEL

MALE FEMALE

PUBLIC

TRANSPORT

2 2

PRIVATE

TRANSPORT

8 8

TOTAL 10 10

18 | P a g e

Page 19: ECO REPORT.docx

GRAPH

PUBLIC TRANSPORT

PRIVATE TRANSPORT

0

1

2

3

4

5

6

7

8

MALE

FEMALE

MALEFEMALE

19 | P a g e

Page 20: ECO REPORT.docx

Q:5 ) If you get 20% decrease in your salary/pocket money you will travel by?

Aim: The Aim of this question is to know that by which mode respondent will travel if there is

20% decrease in salary/Pocket money.

Response Format: It is a Multiple choice question.

No of respondent to answer this question: 20

No of answers generated: 20

TABULATED RESULT

AGE MALE FEMALE

PUBLIC

TRANSPORT

8 6

PRIVATE

TRANSPORT

2 4

TOTAL 10 10

20 | P a g e

Page 21: ECO REPORT.docx

GRAPH

PUBLIC TRANSPORT PRIVATE TRANSPORT0

1

2

3

4

5

6

7

8

MALEFEMALE

21 | P a g e

Page 22: ECO REPORT.docx

CONCLUSION AND FINDING

In this research we had found that when there is increase in the income of the person it more

towards the superior good like in our case when there is an increase of about 20% the person start

using its private transport as the mode of transportation but when there is a decrease in its income it

starts using public transport as the mode of transportation as it is cheaper than Public transport.

This thing proves that there is the effect of income elasticity on Private Transport

22 | P a g e

Page 23: ECO REPORT.docx

BIBLIOGRAPHY

1. Frankena, M. . Income Distributional Effects of Urban Transit Subsidies. Journal of

Transport Economics and Policy, 7(3), 215-230

2. http://www.tutor2u.net/economics/revision-notes/as-markets-income-elasticity-of-

demand.html

3. http://www.econport.org/content/handbook/Elasticity/Income-Elasticity.html

4. http://www.economicsonline.co.uk/Competitive_markets/

Income_elasticity_of_demand.html

5. http://moneyterms.co.uk/income-elasticity/

6. http://www.extension.iastate.edu/agdm/wholefarm/pdf/c5-207.pdf

7. Karl E. Case and Ray C. Fair, Principles of Economics, 

8. Samuelson, Paul A. and Nordhaus, William D.,  Economics ,  

23 | P a g e