Evidence from Puno, Peru

Transcripción

Barrantes
Impact of mobile phone on agricultural profits from livestock activities
The impact of mobile phones on profits from livestock
activities – Evidence from Puno, Peru
Roxana Barrantes
Instituto de Estudios Peruanos
[email protected]
BIOGRAPHY
PhD-University of Illinois at Urbana-Champaign. Currently, Principal Researcher at Instituto de Estudios Peruanos (IEP),
and Associate Professor, Department of Economics, Pontificia Universidad Católica del Perú. She is member of the Steering
Committee of DIRSI (Regional Dialogue for the Information Society) and member of the Scientific Committee of the
PICTURE-Africa Research Project.
ABSTRACT
Besides the work of Jensen (2007), there is little quantitative evidence on the impact that mobile telephony has had on
household welfare. In considering the rural household welfare, the possibility is open of finding impacts of information that is
accessed via mobile phone in several markets where rural households are usually inserted: agricultural product markets,
agricultural services markets, agricultural byproducts; but also in labor markets that often supplement income diversification
strategies of these households. Using a database collected to measure the impact of mobile telephony in the welfare of rural
households in Puno, Peru, this paper seeks to focus attention on the markets for agricultural products and by-products. The
aim is to measure the contribution that has the use of mobile telephony in the profits resulting from the development of
agricultural activities, using econometric techniques associated with quasi-experimental methods of impact assessment. How
much does the mobile phone contribute to agricultural earnings? What is the differential impact of mobile phone use vis-a-vis
scale variables such as farm size or the number of cattle, or diversification, as the total number of crops, or vertical
integration, as the production of agricultural products, on the results of farming? We expect to find different impacts
depending on the type of use of mobile telephony, ie if used for information to affect the agricultural production function or
is used to make marketing decisions. The results can help justify public policy efforts to include mobile telephone service as
a basic service as well as the development of specific mobile livelihood services for farmers from the mobile communication
technology, yet absent in Latin America.
Keywords (Required)
Mobile phone use, agriculture, rural areas Latin America, Peru
1. INTRODUCTION
In less-developed countries, mobile phones are the preferred means of access to telecommunications services, particularly
among the poor, who show different strategies that combine mobile phones to receive calls and public telephones to make
calls (Galperin and Mariscal (2007), Barrantes (2007), Gutierrez y Gamboa (2007), Ramírez and De Angoitia (2008), among
others). In rural areas, which usually lack fixed telephony and public phones, there was a delay in the expansion and,
therefore, adoption of mobile phone service. In addition, poverty is concentrated in rural areas, making them unattractive for
commercial service expansion. Despite these difficulties, mobile phones are widely used in rural areas, although subscription
to pre-paid phones lags behind use, and post-paid service is almost non-existent. The discrepancy between use and
subscription is partly explained by the widespread availability of mobile call services offered by street vendors; this service is
essentially a substitute for public phones.
Using quantitative data gathered in the area of influence of two rural markets in Puno, in southern Peru, where livestock
raising is as important as crop farming, this paper aims to identify the contribution of mobile phone use to profits derived
from agricultural activities. The impact of “directly productive” uses, such as communicating with clients, suppliers or
Proceedings of the 4th ACORN-REDECOM Conference Brasilia, D.F., May 14-15th, 2010
1
Barrantes
producers’ associations, on agricultural profits is identified. Based on previous work (Barrantes, Agüero, Fernández-Ardevol,
2009) which examined the effect of mobile phone use on household welfare, this paper focuses on the productive side of the
agricultural household, and does not consider the possible contribution to family welfare of the inclusion of household
members in labor markets. This paper builds upon Barrantes (2010) by focusing on mobile phone users and refining the
econometrics for households whose main activity is livestock husbrandry.
The evidence shows a strong effect of mobile phone use on profits from livestock, that does not extend to explaining profits
from crop farming. Moreover, the distinction introduced in this paper between mobile phone use for obtaining information
relevant for the production function and the information needed to marketing decision making is proved to be significant in
the case of livestock husbandry. As expected, variables such as the household’s commercial orientation or the vertical
integration of the production process are also important in explaining the level of profits attained. Because mobile phone use
is very recent for these producers, the median length of use being 12 months, information relevant to production processes
that is gathered by using the mobile phone does not yet have a significant impact on crop production and does not have the
expected effect on livestock production.
The structure of the paper is as follows. This introduction is followed by a brief description of the study area. The next
section describes the analytical framework. Econometric results are presented in the fourth section. The paper ends with final
comments and pending research questions.
2. DESCRIPTION OF THE STUDY ZONE
The information used in this study was collected in June and July 2008 as part of the study of “Mobile Communications and
Development in Latin America,” funded by the Fundación Telefónica and led by the Universitat Oberta de Catalunya (UOC).
A random sample of homes was chosen in the areas of influence of two markets in the Puno region, in southern Peru 1 , to
evaluate the impact of the introduction of mobile telephones on daily life in rural homes. One person between ages 13 and 70
was randomly chosen from each household to learn about mobile phone use. This informant was given an additional
questionnaire about the use of mobile phones and other ICTs in general.
The markets were chosen controlling for similar key characteristics: altitude and population. Altitude is a very important
geographical constraint in the area of the Collao Plateau, which is part of the Lake Titicaca ecosystem. 2 Local altitudes on the
plateau exceed 3,500 meters above sea level. Unlike the rest of the Peruvian Andes, it is basically flat, with few of the steep
slopes that make productive activity difficult. Although the slopes are relatively gentle, households in this area of Puno face
extreme weather conditions during the day and/or throughout the year. In winter, they suffer ground frost, which hits them
hard and for which they are not prepared. Besides geography, the study looked for similarities in the size of the villages,
measured by number of inhabitants, and the poverty level of the households, using unmet basic needs as the indicator. 3 The
markets were chosen based on those three basic criteria.
The markets chosen were in Asillo and Taraco, in the provinces of Azángaro and Huancané, respectively. From Juliaca, the
commercial capital of Puno, it takes about an hour to reach either of them on a paved road.4 Asillo’s market day is Sunday,
while the Taraco market is on Thursday. Both are held from 5 a.m. to 3 p.m. Six districts were identified in the Asillo market
area and 10 in the area near the Taraco market.
Table 2.1 shows the poverty indicator based on the number of unmet basic needs (UBN) for the households in the sample
surveyed for the qualitative study. Three of every four households have at least one UBN, which is well above the nationallevel indicator.
Table 2.1 Unsatisfied Basic Needs (UBN) in the study area
Indicator
No UBN
1
A map can be found in the annex.
2
See Parodi (1995).
3
See Feres y Mancero (2001).
Sample*
24%
Asillo*
20%
Taraco*
27%
Puno**
26%
Peru**
41%
4
Both villages can be reached from Lima via Juliaca (San Román province), which has an airport. The flight takes about an
hour and a half. Once in Juliaca, visitors can travel to Asillo by public transportation (bus). The fare is S/.4.00 Sol (US$ 1.30)
and the journey takes about two hours. Visitors can travel to Taraco from Juliaca by rural vans (called “combis”), a trip that
takes about 45 minutes and costs S/.2.50 (US$0.83).
2
Barrantes
1 UBN
2 UBN
3 UBN
4 UNB
5 UBN
33%
29%
11%
3%
0%
33%
28%
15%
4%
0%
32%
30%
8%
3%
0%
20%
24%
18%
10%
1%
19%
18%
14%
8%
1%
Source: * Survey (Barrantes, 2008) and ** ENAHO (National Living Standards Survey) 2007, for Puno and Peru.
3. FRAMEWORK FOR ANALYSIS
It is widely recognized that in developing countries, mobile telephony, chiefly for low-income sectors and rural areas, has
given people their first opportunity to access telecommunications. When people use mobile telephones, they obtain
information and lower the costs of communicating, helping them establish more solid positions in markets, gain access to
new markets, and increase their income by reducing losses from price dispersion.
Jensen (2007) conducted the study that has had the greatest impact on knowledge of the effects of mobile telephones, by
demonstrating that rent dissipation caused by incomplete information is reduced by using mobile telephones, which supports
the law of a single price and the efficient working of markets, in the context of fresh fish markets in Kerala, India. Similarly,
with evidence collected in Niger, Aker (2008) found that the use of mobile telephones reduced price dispersion in the grain
market; the decrease was greater in more remote markets with less access. It is important to note that these two studies focus
on the role of the information mobile telephones provide in marketing activities, not in those related to what economists will
call the production function.
Esselaarc et al. (2007) studied the impact of ICTs in small businesses and microenterprises in 13 countries in Africa. The
main finding was that these technologies are highly productive inputs, because they reduce transaction costs and provide
greater market access both for the formal and informal sectors. They stress the use of mobile phones, reporting an immediate
benefit because they are easy to use and are widely available.
As in other research (Galperin and Mariscal, 2007; De Silva and Zainuden, 2007), this study distinguished between the owner
of the telephone (subscriber) and the user. Due to affordability constraints, the user may not necessarily be the subscriber. In
fact, survey figures show that 76 percent of interviewees are service users, and of these, just two-thirds are subscribers. The
mobile telephone is shared by members of one family or by various friends. There is also a considerable supply of calls
through mobile phones for public use, by street vendors or chalequeros who offer the service, or through phone booths or
telecenters. 5
This study begins with a simple household production function model, to explain not the level of production, but the level of
profit from livestock and crop farming. While the interaction between those activities is recognized, this study separates the
estimated profit from crops from the profit from livestock. In each case, direct sales and by-products are added. While the
former constitute a primary activity, the latter represent processing, postulated to give greater added value to primary
production.
Profit (the difference between revenue and costs) is therefore due to two factors: production and marketing. In the area of
production, I argue that profit depends on the level of certain stocks of human and natural capital. Marketing management is
also the outcome of decisions linked to stock flows, reflected in the degree of insertion in markets. Besides these variables,
which are typically discussed in the literature, and which explain small farmers’ production decisions and outcomes, this
study also includes characteristics and perceptions of the use of mobile phones for obtaining information for production and
marketing decisions. The variables, their definitions and the underlying hypothesis are summarized in Table 3.1.
The variables chosen to reflect human capital stock are: total size, indicated by the number of household members; the
proportion of adults, which reflects the importance of the most productive labor; and accumulated human capital, based on
the educational level of the household member with most years of schooling.
5
Chalequeros are people who hire out mobile telephones by the minute. They usually work in village squares or on busy
street corners and they wear bright-colored vests (hence the name, which comes from the Spanish word for vest, chaleco).
Their rates are lower than public telephone and pre-paid phone rates.
3
Barrantes
I also consider variables associated with the use of the mobile telephone as a production input:using the mobile phone to
communicate with clients, suppliers or producers belonging to associations, which indicates a connection with marketing
decisions; using the mobile phone to obtain information about crop or livestock production, which indicates a connection
with production decisions; or a perception that communication has improved with the use of the mobile telephone.
The models include a dummy variable that places the household in the area of influence of a particular market, with Taraco
having a value of zero.
In the case of natural capital, crop farming is distinguished from livestock raising. For crop farming, the model considers
average farm plot size, which indicates the possibility of achieving economies of scale in production; the number of plots,
which reflects both a possible strategy for reducing climate risks and the division of land, which is an obstacle to the
increases in efficiency that are possible with a higher productive scale; and the number of crops, which shows crop diversity
and risk reduction, as well as a lack of specialization, which can negatively affect profit.
Market orientation and production results are measured by various ratios. First, as an indicator of the importance of primary
activities, is the relative importance of crop sales in total sales. Second is the relative importance of crop production for
making agricultural by-products, which shows vertical integration; and the proportion of fodder crops in total agricultural
production. The third factor is the importance of the main crop as an indicator of specialization and possible associated
efficiencies.
The analysis of livestock husbrandry differs from that of crop farming in the definition of natural capital variables and the
ratios that reflect market orientation. As natural capital variables, the study considers the number of species of animals, which
is an indicator of diversification and risk reduction, but which is also an obstacle to obtaining the benefits of specialization;
the number of heads of the most valuable kind of animals, as an indicator of productive specialization; and average pasture
size. The variables used to analyze market insertion reflect the relative importance of certain types of production: livestock
value compared to total added value, as an indicator of the importance of primary activities; the value of the main species
compared to the total for all livestock, as an indicator of specialization; the importance of fodder crops; and the value of the
main by-product as a percentage of all by-products.
Table 3.1: Variables included in the econometric analysis
Variable
Indicator
Agricultural profit
Type Measurement unit
Definition / Hypothesis
Continuous
(Current Soles)
Agricultural profit is he difference
between revenue and total agricultural
expenditure.
Agricultural income is the sum of the
total value of agricultural production,
the value of agricultural by-products
and the total value of forest
production. Agricultural expenditure is
the sum of wages, animal and machine
hiring and other inputs.
Continuous
(Current Soles)
Livestock profit is the difference
between revenue and total livestock
expenditure.
Livestock income is the sum of the
value of revenue from livestock
activity (sale of animals) and the total
value of livestock by-products.
Endogenous
variable
Livestock profit
Human capital
Number of household
members
Discrete
The value of this variable is the total
number of members in each
household. A higher value is related
to higher profits, because it minimizes
the need to hire labor.
4
Barrantes
Proportion of adults en in
household
Continuous
(Real number between 0 and
1)
This ratio is the quotient of adults per
household (between ages 15 and 65)
divided by the total number of
household members.
A higher ratio means higher profit,
reflecting a more productive labor
force.
Highest level of education
achieved by a household
member
Discrete
Whole number
A household member with more
education can have a positive impact
on productivity.
Number of plots
Discrete
Whole number
The value is the number of plots of the
household.
A higher number may be related to
land fragmentation, which results in
difficulties in achieving economies of
scale. It could therefore be associated
with low productivity, which
negatively affects the level of
agricultural profit.
Average plot size
Continuous
(Hectares)
A smaller average plot size may
adversely affect productivity and thus
the level of agricultural profit
Discrete
(Whole number)
A larger number of crops in the
portfolio is expected to be associated
with lower levels of agricultural
income and difficulties in
specialization, which makes it more
difficult to achieve economies of scale.
Natural capital Agriculture
Number of crops
Number of species
Discrete
Whole number
Natural Capital –
Livestock
production
Productive results
and market
orientation Agriculture
The number of species of animals
raised by the household.
A higher number of species shows
greater diversification and thus a
reduced risk, which can have a
positive impact on livestock profit
level.
Number of head of the main
animal
Discrete
Whole number
The main animal is the one that
contributes the greatest added value
associated with livestock.
A larger number of animals is
expected to be associated with greater
livestock profit.
Value of production of
fodder crops per hectare
tilled (including own and
rented)
Continuous
(Current Soles)
Higher value implies greater
productivity, and greater agricultural
profit is therefore expected.
Ratio: Value of fodder crops
/ Total value of agricultural
production
Continuous
1)
For greater integration of crops and
livestock, the importance of fodder
crops may reflect vertical integration
and be associated with higher profit
levels.
Ratio: Value of production
Continuous
This ratio reflects the importance of
5
Barrantes
Production
outcomes and
market orientation
– livestock
production
devoted to making
agricultural by-products /
Total value of agricultural
production
(between 0 and 1)
vertical integration and may reflect
higher profit levels.
Ratio: Value of main crop /
Total value of agricultural
production
Continuous
(between 0 and 1)
Indicates greater specialization and is
related to higher productivity and
profit.
Livestock raising household
Dichotomous
= 1 if agricultural household
If a crop-farming household also raises
livestock, the result could be risk
reduction through diversification, but
also higher diversification and
difficulties in achieving economies of
scale.
Value of fodder production
per hectare tilled (including
own and rented)
Continuous
(Current Soles)
Higher value implies greater
productivity, so higher livestock profit
is expected.
Importance of by-product
sales compared to total
added value of livestock
production
Continuous
1)
This ratio shows the relative weight of
livestock by-products in the total
added value. This may be related to
greater productivity and thus be
associated with higher levels of
livestock profit.
Importance of main byproduct sales compared to
total livestock by-product
sales
Continuous
(From 0 to 1)
Indicates greater specialization and is
related to greater livestock
productivity and profit.
Dichotomous
= 1 if mobile was used for
that purpose.
Multiplicative variable that establishes
interaction between the variable “use
of information from third parties for
agricultural production” and the
variable “use of mobile phone for
obtaining information”.
Dummy – used mobile to
get information about …
--either agricultural crops or
livestock production
Continuous
(months)
Length of time mobile
phone has been used
Mobile phone as
production input
Categorical
Under 1 year.
From 12 to 24 months
Over 24 months
Having used a mobile phone for a
longer time reflects greater familiarity
with it and knowledge of its use. This
may help in obtaining information.
Dichotomous
= 1 if mobile was used for
that purpose.
The variable considers the informant
who uses the mobile phone to
communicate with clients and/or
suppliers and/or members of producers
associations or cooperatives.
Decreased transaction costs can have a
positive effect on the levels of profits.
In the OLS models, only
communication with clients or
suppliers is considered. The IV models
add communication with members of
producers’ associations or public
6
Used mobile phone to
communicate with clients,
suppliers or members of
producers’ associations
(dummy)
Barrantes
agencies.
Location
Dummy if perceived
improvement in
communication
Dichotomous =1 if
communication is perceived
to have improved a little or
greatly
If the informant perceives
improvement in communication, this
may signal full integration of the
mobile phone into everyday activities.
Market
Dichotomous
= 1 if the market is in Asillo.
A location variable.
4. EMPIRICAL ANALYSIS
The empirical strategy is to explain the level of earnings in the respective activity (crop or livestock). The level of profits can
be explained or per capita household level. Unlike Barrantes (2010), where the emphasis was placed on comparing users
versus non-mobile users, this paper emphasizes the different potential uses of mobile phones both in the field of marketing
decisions and of the production sphere as well. Hence the analysis is restricted to agricultural households where the informant
is a user and also is the head of household or spouse.
The emphasis is thus placed in elucidating the role of using the mobile phone in decisions related to agricultural production in
the dimensions affecting the production function. It is postulated that the mobile phone plays a role as a productive input,
when it allows to access information more cheaply and timely than other ICT. The effect of using this information is different
when it pertains to aspects related to the production function, ie the combination of inputs to produce, as compared to those
aspects related to marketing, ie, decisions of the time and place of sale. Consequently, the effect of mobile phone use will be
different if it involves decisions on production or on marketing decisions.
To account for the varied possible productive uses of the mobile phone, several indicators were used as regressors: whether
the informant used the mobile phone to get information for the production process (agricultural or livestock), whether
information obtained from family members was used in production combined with whether the mobile phone was used to
communicate with family; and if the informant used the mobile phone to communicate with customers, suppliers, similar
businesses, association, cooperative, or any support institution. The first two correspond to uses that would affect the
production function and the last indicator reflects mobile use to affect marketing decisions.
Obviously, agricultural production or livestock production or the respective by-products, depend on other inputs as well as
other controls - such as, for example, the location of the fair. The variables and indicators used can be found in Table 3.1, and
were explained in the previous section.
It is important to stop and explain a key element of the empirical strategy, which is the use of instrumental variables. It seeks
to unravel the problem of causality involved when the mobile phone is used as an explanatory variable of the level of profits,
as it reflects the access to information as a productive input, when it could well be that the level of earnings accounts for the
highest probability of using the mobile as productive input, as communication is key to successfully penetrate markets. Then,
using the 2SLS procedure, the productive use of mobile phones is instrumented with three variables: being a subscriber, to be
a user for a longer period of time, and to perceive a higher quality of service.
The database contained information from households that stated that their permanent activity was crop farming (699) and
those that said they were dedicated to raising livestock (690). There could be some overlap, because the two activities tend to
be complementary for rural families (667 households). However, since the goal was to identify the impact of mobile use as a
productive input, the regression analysis only included households where the informant was a mobile phone user. Therefore,
the total number of households for each kind of activity shrank: from 699 to 427 for agriculture, and from 690 to 393, in for
raising livestock.
Similarly, households are grouped by main crop, or by the most important type of herd, or main byproduct. The hypothesis to
justify this strategy rests on the different productive cycles and marketing of various products, which can be more clearly
appreciated when isolated regressions are run. The descriptive statistics for all variables used can be found in Appendix 1.
4.1. Profit from raising livestock
The set of regressions explaining the level of profits attained from raising livestock –be it total level or per capita— can be
found in Table 4.1. Models 1 and 2 consider all households, while regressions 3, 4, and 5 consider households which raise
vacuno criollo, and regression 6 is run on milk producers –what is considered a by product of livestock raising.
7
Barrantes
The results of the regressions, run on the natural logarithm of the dependent variable, are shown in Table 4.1., and reflect an
appropriate overall goodness of fit for all models. Further tests were run on both coefficient bias, and instrument strength,
yielding acceptable results. 6
The use of mobile phones to communicate with clients, suppliers and members of producers’ associations shows the expected
positive sign and is statistically significant in all models. On the other hand, the use of mobile phones to gather information to
decide on productive aspects of raising livestock show a negative sign and is statistically significant only when all raising
livestock households are considered. The effect of mobile use runs in opposite directions in the sample: positive for
communications for marketing and negative for directly productive use –those affecting the production function.
Insertion in fodder markets and specialization, reflected in the relative importance of the main by-product in total value
added, are statistically significant and show the expected positive sign. Livestock size also positively influences the level of
profits. Livestock profits are not affected by market location, as shown by the coefficient on Fair.
In Model 2, variables of scale of production (number of most important animal heads) and the variables that indicate vertical
integration (fodder crops and the importance of by-products in total livestock production), are also significant. In the latter
case, it is interesting that the greater the importance of byproducts, the smaller the profits from raising livestock, indicating an
internal subsidy. Human capital variables are not significant in any of the models.
6
Shown in Table 4.1 by the Cragg-Donald statistic and the F-Test for excluded instruments.
8
Barrantes
Table 4.1. Regression Results
Livestock HH
Dependent variable (ln)
Explanatory variables
Mobile phone used to communicate
with clients and suppliers, similar
businesses, producers’ associations or
support agencies
Mobile phone used
information
about
production
to
Model 1
Per capita profits
when respondent is
user and head of
household or spouse
0,4406106 *
(0,2458339)
Vacuno criollo HH
Model 2
Profits when
respondent is user
and head of
household or spouse
Model 3
Per capita profits when
respondent is user and
head of household or
spouse
Milk producers
Model 4
Model 5
Profits when respondent is user and head of
household or spouse
0,5798816 **
(0,2684291)
Model 6
Profits when
respondent is user
and head of
household or spouse
0,457231 **
(0,2592699)
0,6686473 *
(0,271398)
0,571353 **
(0,2567895)
0,7134942 **
(0,2871292)
-0,111385 **
(0,0656013)
-0,1182095
(0,078197)
-0,105943
(0,0724004)
-0,1055344
(0,073036)
-0,1080199
(0,0830563)
0,0122159
(0,0099088)
0,012208
(0,0095558)
-0,0028791
(0,0092092)
0,0128619
(0,1153263)
-0,0679248
(0,1082217)
obtain
livestock
-0,1226931
(0,0279835)
**
Highest level of education reached by
a member of household
-0,0001607
(0,0085812)
Share of adults in household
Ratio: Total value of livestock byproducts/Total added value of
livestock production
Ratio: Sales value of main byproduct/total value of livestock byproducts
Value fodder production per hectare
tilled (includes own and rented)
Number of species
Number of heads of main animal
-0,0441548
(0,1033706)
-0,2516858 ***
(0,0789144)
-0,2075331 ***
(0,0757046)
-0,4251128 ***
(0,1120549)
-0,3508229 **
(0,0978637)
-0,3599909 ***
(0,1032908)
-0,6868591 ***
(0,0972016)
0,3066641 ***
(0,0637117)
0,2949658 ***
(0,0608551)
0,3650294 ***
(0,0770273)
0,3333376 **
(0,07137)
0,3317105 ***
(0,0714842)
0,2131643 ***
(0,0625804)
0,0000742 **
(0,0000296)
0,0000681 **
(0,0000277)
0,0000488
(0,0000299)
0,0000432 *
(0,0000243)
0,0000469 *
(0,0000275)
0,0000495 *
(0,0000267)
0,1244187 ***
(0,0279835)
0,1549703 ***
(0,0273239)
0,0389867
(0,0415321)
0,0743231 *
(0,0401105)
0,0750054 *
(0,0402174)
0,0348474
(0,0321752)
0,0001586 ***
0,0001264 **
0,0001214 **
9
Barrantes
(0,0000498)
(0,0000532)
(0,0000487)
Fair (Asillo = 1)
0,1231607 *
(0,0676947)
0,096161
(0,0620557)
0,0425908
(0,0748523)
Constant
6,726398 ***
(0,0800741)
7,892213 ***
(0,1173034)
7,023034 ***
(0,1291656)
Goodness of Fit Statistics
Number of observations
Degrees of freedom
Cragg-Donald Statistic
F-test for overall instruments
Centered R2
Uncentered R2
0,0229301
(0,069978)
8,032066 ***
(0,1280472)
0,135 *
(0,0711188)
8,01232 ***
(0,1438166)
8,555354 ***
(0,1455535)
393
7
393
10
253
7
253
8
253
10
294
9
11,463
2,21 *
0,1113
9,377
2,80 **
0,1569
8,699
2,35 *
0,0814
7,438
3,07 **
0,1587
7,041
2,97 **
0,1554
6.480
2.76 **
0.1008
0,9957
0,9972
0,9956
0,9973
0,9973
0.9975
Instruments: Length of use of mobile, perception of improved quality, terminal owner
Stock-Yoho Critical Values:
5% maximal IV relative bias
13,91
9,08
6,46
30% maximal IV size
5,39
10% maximal IV size
15% maximal IV size
20% maximal IV size
25% maximal IV size
Standard errors in parenthesis
*** Significance level = 0,01
** Significance level = 0,05
* Significance level = 0,10
10
22,30
12,83
9,54
7,80
Barrantes
Impact of mobile phone on agricultural profits from livestock
activities
4.2. Crop farming profit
The hypothesis is that the level of profit from crop farming, either total or per plot, depends on the levels of human capital
and natural capital stock, the degree of specialization and market orientation, and the use of mobile telephones to facilitate
access to information and reduce overall transaction costs.
None of the models showed statistically significant results for any of our variables indicating mobile phone use.
FINAL COMMENTS
Using quantitative evidence gathered in the area of influence of two rural markets in Puno, in southern Peru, this paper shows
the positive effect of mobile phone use on profits from livestock production in rural households. Higher profits are explained
by the use of mobile phones by heads of households or spouses who make calls to clients, suppliers, similar businesses,
producers’ associations or support agencies.
In contrast to our initial expectation, the econometric results did not extend to agricultural profits. None of the postulated
variables indicating mobile phone use, either for production or marketing decision making, were significant in explaining
agricultural profits.
The underlying hypothesis in the econometric modeling is that the cost of looking for new markets for a particular product is
lower than the cost of adopting new techniques, which may be associated with modification of the product. Information
leading to product modification may take longer to permeate entrenched agricultural practices that have proven to reduce risk
over the years. The possible positive effects of the use of mobile phones on profit of raising livestock, occur first in marketing
and are not yet manifested or perceived in defining parameters for production, that is obtaining information about raising
particular animals or producing by-products.
This possible differentiated effect could mainly be a response to the fact that these decision-makers have used mobile phones
for only a short time, an average of barely over a year -16 months. During that time, they have made many more marketing
decisions about the farm household’s products or by-products than about production (decisions associated with the crop cycle
or animal reproduction cycle). Given this length of use, it may be too early to assess the directly productive impact of mobile
phone use for these rural producers.
Nevertheless, it is important to note the statistically differentiated effects observed when explaining agricultural profits
compared to livestock profits. The latter appear more conclusive than the former. This could be because the production time
frame is more flexible for livestock production than crop farming. The qualitative evidence gathered for the study (Aronés,
León y Barrantes, 2009), showed that timely contact with a veterinarian was key to increasing livestock productivity; this was
achieved by using the mobile phone. No similar key use of the mobile phone was documented for crop production.
On the other hand, emphasizing calls to clients and suppliers as an indicator of the productive use of the mobile telephone
overlooks the fact that these households’ information networks are crisscrossed by solid kinship relations in contexts in which
market transactions have not yet permeated a wide array of activities, as they would in more modern or urban areas of the
country. It is difficult to determine when a call to a relative stops being ‘unproductive’ and becomes a productive call (i.e.,
related to a decision about where to sell, price, inputs, etc.). Clearly this is an area for further investigation.
ACKNOWLEDGMENTS
I would like to thank several IEP young researchers: Ramón Díaz for initial discussions, which helped me define the
approach; Ramiro Burga, who pursued the econometrics; Aileen Agüero, who contributed to the literature review, and Oscar
Madalengoitia, who drafted the map. Comments by Jonathan Donner, Mireia Fernandez-Ardevol and participants at the
Conference on Development and Information Technologies: Mobile Phones and Internet in Latin America and Africa: What
Benefits from the most disadvantaged? held in Barcelona in October 2009, are greatly appreciated. The usual disclaimer
applies.
REFERENCES
1.
Aker, J. (2008) Does digital divide or provide? The impact of cell phones on grain markets in Niger. Berkeley:
University of California. http://are.berkeley.edu/~aker/cell.pdf (20/04/09).
11
Barrantes
activities
2.
Aronés, M., León, L. y Barrantes, R. (2009) La telefonía móvil en el ámbito rural. Estudio cualitativo de las áreas de
influencia de las ferias de Asillo y Taraco, en Puno – Perú. Mobile Communications and Socioeconomic Development in
Latin America Project. Final report. Unpublished manuscript.
3.
Barrantes, R. (2007) Oportunidades móviles: pobreza y acceso a la telefonía en América Latina y el Caribe. El caso de
Perú. Lima, DIRSI. http://dirsi.net/sites/default/files/dirsi_07_MO_per_es.pdf. (09/04/10).
4.
Barrantes, R., (2008) Comunicaciones Móviles y Desarrollo Socioeconómico en América Latina. Módulo II: Proceso
social de Desarrollo. Estudio de caso 2: La telefonía móvil en el ámbito rural. Partial report.
5.
Barrantes, R., (2010) Mobile phones as a tool in the household production process Evidence from Puno, Peru.
Communication Technologies in Latin America and Africa: A multidisciplinary perspective. UOC y Agencia Catalana de
Cooperació al Desenvolupament. Barcelona. Pp. 87-116.
6.
Barrantes, R., Agüero, A. and Fernández-Ardevol, M. (2009) La telefonía móvil en el ámbito rural. Estudio de caso de
los hogares de Puno-Perú. Mobile Communications and Socioeconomic Development in Latin America Project. Final
report. Unpublished manuscript.
7.
De Angoitia, R. y Ramírez, F. (2008) Estrategias utilizadas para minimizar costos por los usuarios de telefonía celular de
sectores
de
bajos
ingresos
de
México.
Lima,
IDRC,
Serie
Investigaciones
breves,
2.
http://dirsi.net/sites/default/files/dirsi_08_RB2_es.pdf. (09/04/10).
8.
De Silva, H. and Zainudeen, A. (2007) Teleuse on a Shoestring: Poverty reduction through telecom access at the ‘Bottom
of the Pyramid. Paper prepared for Centre for Poverty Analysis Annual Symposium on Poverty Research in Sri Lanka.
http://www.lirneasia.net/wp-content/uploads/2007/04/lirneasia_teleuse_cepa_-mar07_v30.pdf (13/01/09)
9.
Donner, J. (2006) The Use of Mobile Phones By Microentrepreneurs in Kigali, Rwanda. Information Technologies and
International Development, 3, 2 – Winter.
10. _____, (2008) Research Approaches to Mobile Use in the Developing World: A Review of the Literature. The
Information Society, 24, 140–159.
11. Esselaarc S., Stork, C., Ndiwalana, A. and Deen-Swarray, M. (2007) ICT Usage and its impact on profitability of SMEs
in 13 African Countries. Information Technologies and International Development, 4, 1, 87–100.
12. Feres, J., Mancero, X. (2001) El método de las necesidades básicas insatisfechas (NBI) y sus aplicaciones en América
Latina. Estudios y perspectivas. ECLAC – United Nations. http://eclac.cl/deype/mecovi/docs/TALLER5/8.pdf.
(15/10/08).
13. Figueroa, A. (1983) La economía campesina de la sierra del Perú. Lima: Pontificia Universidad Católica del Perú.
14. Galperin, H. y Mariscal, J. (2007) Oportunidades Móviles: Pobreza y Telefonía Móvil en América Latina y el Caribe.
DIRSI. http://dirsi.net/sites/default/files/dirsi_07_MO_reg_es_0.pdf. (09/04/10).
15. Gutierrez, L. y Gamboa, L. (2007) Oportunidades móviles: pobreza y acceso a la telefonía en América Latina y el
Caribe. El caso de Colombia. Lima, DIRSI. http://dirsi.net/sites/default/files/dirsi_07_MO_col_es.zip. (09/04/10).
16. Jagun A., Heeks, R., Whalley, J. (2007) Mobile Telephony and Developing Country Micro-Enterprise: A Nigerian Case
Study. Working Paper Series No. 29. Institute for Development Policy and Management.
http://www.sed.manchester.ac.uk/idpm/research/publications/wp/di/documents/di_wp29.pdf (27/04/09).
17. Jensen, R. (2007) The Digital Provide: Information (technology), market performance and welfare in the South Indian
fisheries sector. The Quarterly Journal of Economics, 122, 3, 879-924.
18. Parodi, A. (1995) El lago Titicaca: sus características físicas y sus riquezas naturales, arqueológicas y arquitectónicas.
Arequipa, Regentus.
19. Souter D., Scott, N., Garforth C., Jain R., Mascarenhas O. and McKemey K. (2005) The economic impact of
telecommunications on rural livelihoods and poverty reduction: a study of rural communities in India (Gujarat),
Mozambique and Tanzania. Commonwealth Telecommunications Organisation for UK Department for International
Development.
http://iimahd.ernet.in/ctps/pdf/The%20Economic%20Impact%20of%20Telecommunication%20on%20Rural%20Livelih
oods-Teleafrica%20Report.pdf. (08/05/08).
12
Barrantes
activities
Map 1. Puno and the areas of influence of the Asillo and Taraco markets
13
Barrantes
activities
Appendix 1 – Descriptive Statistics
All monetary figures are expressed in Soles. Current exchange rate: 2.8 soles per American dollar.
Subset: Respondent is user and head of family or spouse, and raising livestock; N=393
Variable
Mean
Median
SD
Min.
Max.
%Yes
%No
ln Profit per cáp
7.14
7.04
0.50
6.35
9.31
ln Profit
8.36
8.30
0.48
7.57
10.42
0.43
0.38
0.34
0.00
1.00
10.75
12.00
3.26
1.00
16.00
Ratio adults HH
0.62
0.60
0.25
0.00
1.00
Relative importance of main byproduct
0.43
0.43
0.43
0.00
1.00
751.34
180.00
1229.61
0.00
7045.46
2.37
2.00
0.94
1.00
6.00
16.36
2.00
252.20
0.00
5000.00
Market
0.55
1.00
0.50
0.00
1.00
55%
45%
Mobile-intra-fam-info
0.80
1.00
0.40
0.00
1.00
80%
20%
Quality perception
0.70
1.00
0.46
0.00
1.00
70%
30%
Terminal owner
0.64
1.00
0.48
0.00
1.00
64%
36%
16.40
12.00
15.38
1.00
120.00
0.15
0.00
0.36
0.00
1.00
15%
85%
Relative
products
importance
of
by-
Max edu HH
Agri-livestock VI per ha.
Number of species
Size of main species
Length of use
Mobile-extra-familiar
14
Barrantes
activities
Subset: Respondent is mobile phone user and head of household or spouse, raising vacuno criollo; N=253
Variable
Mean
Median
SD
Min.
Max.
%Yes %No
ln Profit per cáp
7.17
7.11
0.50
6.35
9.31
ln Profit
8.39
8.32
0.48
7.57
10.42
0.40
0.35
0.34
0.00
1.00
10.60
12.00
3.31
1.00
16.00
Ratio adults HH
0.63
0.60
0.24
0.00
1.00
0.36
0.00
0.42
0.00
1.00
1049.21
350.00
1425.89
0.00
7045.46
2.50
2.00
0.92
1.00
6.00
23.38
2.00
314.27
0.00
5000.00
Market
0.39
0.00
0.49
0.00
1.00
39%
61%
0.73
1.00
0.45
0.00
1.00
73%
27%
Quality perception
0.78
1.00
0.41
0.00
1.00
78%
22%
Terminal owner
0.61
1.00
0.49
0.00
1.00
61%
39%
16.91
12.00
16.64
1.00
120.00
0.16
0.00
0.37
0.00
1.00
16%
84%
Relative importance
products
of
by-
Max edu HH
Number of species
Length of use
15
Barrantes
activities
Subset: Respondent is Mobile phone user and head of household or spouse, and milk producers; N=294
Variable
Mean
Median
SD
Min.
Max.
%Yes %No
ln Profit per cáp
7.23
7.15
0.48
6.35
9.05
ln Profit
8.45
8.38
0.45
7.61
10.16
0.54
0.48
0.30
0.00
1.00
10.69
12.00
3.26
1.00
16.00
Ratio adults HH
0.62
0.60
0.25
0.00
1.00
0.52
0.74
0.42
0.00
1.00
749.85
200.00
1229.92
0.00
7045.46
2.52
2.00
0.90
1.00
6.00
20.31
2.00
291.54
0.00
5000.00
Market
0.62
1.00
0.49
0.00
1.00
63%
37%
0.84
1.00
0.37
0.00
1.00
84%
16%
Quality perception
0.69
1.00
0.46
0.00
1.00
70%
30%
Terminal owner
0.60
1.00
0.49
0.00
1.00
60%
40%
16.22
12.00
15.60
1.00
120.00
0.16
0.00
0.36
0.00
1.00
16%
84%
Relative
products
importance
of
by-
Max edu HH
Number of species
Length of use
16

Evidence from Puno, Peru

Transcripción

Documentos relacionados

Outlook for US Agricultural Trade. USDA

Agricultural Policy and Trade Reform: The Impact on World