FAO/PARIS21 REGIONAL WORKSHOP ON THE
INTEGRATION OF AND ACCESS
TO AGRICULTURAL STATISTICS FOR
BETTER FORMULATION AND MONITORING OF RURAL
DEVELOPMENT POLICIES
Algers, Algeria 8-9 December 2007
Back-to-back with the 20th AFCAS
AW-07-02-2-A
The Philippine Experience in Establishing Metadata
and Setting up the CountrySTAT
Romeo S. Recide
Maura S. Lizarondo
Jing B. Jalisan
Director, Assistant Director and Senior Statistician, respectively, Bureau of Agricultural
Statistics, Department of Agriculture, Republic of the Philippines
Philippines
1
Structure of Presentation
1. Introduction
2. The National Statistical System
3. Statistical Frameworks
4. Current Approaches in the Maintenance of
Economic Accounts for Food and Agriculture
5. Data Bases in Support of the SEAFA
6. Advantages of the SEAFA
7. Implications of the SEAFA on the Agricultural
Statistical System
8. Improving the Agricultural Statistical Service
9. Conclusions
2
1. Background



The CountrySTAT is an Internet-based system that integrates national food and
agricultural statistical information to ensure harmonization of national data and metadata
collections for analysis and policy making. It provides statistical standards, data
exchange tools and related methods without using external data sources such as
databases. The data source is a text file in a specific format, called PX-file. The
development of the CountrySTAT system started in May 2004 using PX-Web at the FAO
headquarters in Rome. In 2005, the system was successfully tested in the statistical
offices of Kyrgyz Republic, Kenya, and Ghana.
In recognition of the existence and potential uses of the CountrySTAT, the Philippine
Statistical System (PSS) through the Philippine Statistical Association (PSA) has taken
on the challenge of establishing the CountrySTAT system in the Philippines. This has
been made possible through the Letter Of Agreement between the Food and Agriculture
Organization of the United Nations (FAO) and the PSA to implement the project
“Strengthening the National Statistical Systems of Selected Countries in the Asia and
Pacific Region”. The key components of the project were the establishment of the
CountrySTAT Philippines and its introduction to other countries in Asia, the conduct of
the sub-regional workshop on CountrySTAT and Metadata and the preparation and
publication of the Metadata for the national agricultural statistics in 16 Asia-Pacific
countries.
Upon consultation with the key agencies in the PSS, the PSA designated the Bureau of
Agricultural Statistics (BAS) as the host institution for the CountrySTAT. Collection and
integration of data did not pose problems for the BAS since it handles most of the data
requirements of the CountrySTAT. The Statistical Research and Training Center (SRTC)
which is the training arm of the PSS has been tasked to organize and conduct all required
trainings and workshops under the project.
3
2. The CountrySTAT System
 The CountrySTAT is an applied Internet-based information
system that disseminates statistical information and metadata
on food and agriculture. It collects, organizes, integrates and
harmonizes data and metadata from different statistical
agencies in order for statistical tables to be consistent and
compatible at the national, as well as, at the international
levels. The CountrySTAT system facilitates access of
statistical information in a single source which can provide
efficient support to analysis and policy making. In
understanding the CountrySTAT system, one should look at
the developmental framework on which it is based and the
technology that supports and is used in its development.
4
3. Capacity Building for the
CountrySTAT System


3.1. Training on CountrySTAT administration
As an initial step of the capacity building towards the development of
the CountrySTAT Philippines, the PSA sent two BAS personnel to a
training workshop on CountrySTAT Administration at the FAO
Headquarters in Rome, during the period 01 to 10 February 2006.
Earlier in November, one personnel from the National Statistical
Coordination Board (NSCB) attended a similar training at the FAO
HQs. One BAS personnel focused on the CountrySTAT statistical
framework while the other was engaged more on the development,
management and administration of the CountrySTAT system. Apart
from the concepts and the skills acquired about the CountrySTAT
system, the other significant output of this training was the
identification of the requirements and other considerations to enable
BAS to host the CountrySTAT Philippines. This paved the way for
BAS and PSA to proceed and lay out the plans covering the main
activities of the project like; capacity building, conduct of subregional workshop on CountrySTAT and metadata and the
installation of the CountrySTAT at the BAS.
5
3. Capacity Building for the
CountrySTAT System

3.2 National Training Workshops

In the Philippines, the institutions or agencies that supply data to the CountrySTAT system are the
National Statistics Office (NSO), National Statistical Coordination Board (NSCB), the Bureau of Labor
and Employment Statistics (BLES) and the different operating units of the Bureau of Agricultural
Statistics (BAS). The second stage of the capacity building was the training of selected personnel from
these agencies. The project plan scheduled two national training workshops.
The first national training workshop was conducted on 27 to 31 March 2006. The main objective of the
training was to provide and enhance the skills of the statisticians and IT staff in using the PC-Axis
software suite. The three (3) personnel from NSCB and BAS who have earlier attended the training in
Rome served as the resource persons. During the preparations for this training, it has become evident
that the resource persons had to study the PC-Axis software suite in detail in order to have a better
understanding of the PC-Axis. In the process, the resource persons have discovered other aspects of
the PC-Axis and have got to know more than what they have learned from the training in FAO HQs in
Rome.
At the end of the first training, the agencies identified other key elements or requirements for
CountrySTAT. Since data come from different sources, CountrySTAT needs a document that will serve
as a guide to harmonize the data. Specifically, the document should contain guidelines for (1) data
submission, (2) standardized file-naming convention, (3) conversion procedure, (4) standardized format
for PX file keywords. The guidelines will ensure the consistency of statistical tables when viewed in the
CountrySTAT webpage. The CountrySTAT document containing guidelines was drafted by the BAS
Webmaster who serves as the designated CountrySTAT administrator with the help of the participants.



.
6
3. Capacity Building for the
CountrySTAT System

3.2 National Training Workshops

Another important requirement was the forging of a Memorandum of Agreement (MOA)
among agencies involved. The MOA defined the responsibilities and the level of data to
be supplied by the member agencies. This was drafted by SRTC in consultation with the
PSA, NSO, NSCB, BLES and BAS. The MOA underwent several modifications until it was
signed by the heads of agencies on August 2006.
The second national training workshop was conducted during the period 08 to 12 May
2006. The first two days from May 8 to 9 served as a follow-up to the first training and had
the intention to foster better appreciation and understanding in handling PC-Axis files.
This was attended by the participants of the first training workshop. In the next three (3)
days, 10 to 12 May 2006, the training workshop on metadata among Statisticians or
Sector Specialists was held. The workshop included presentation of agencies’ data and
metadata systems. It was interesting to note that none of the agencies have similar
metadata format or outline. The major output of the workshop was the draft metadata
outline. This metadata outline was further enhanced at the BAS. It was submitted to FAO
for information and reference. The draft outline was also circulated among the operating
units of the BAS as reference in the preparation of their respective metadata. Thus, when
the FAO – prescribed metadata outline was forwarded to the BAS, concerned operating
units just did some adjustments in the sequence of presentation of topics.

7
3. Capacity Building for the
CountrySTAT System




3.3 Sub-regional Workshop on CountrySTAT and Metadata
The FAO-PSA Sub-Regional Workshop on CountrySTAT and Metadata, co-organized by BAS
and SRTC, was held in the Philippines from 23-27 October 2006. Basically, the workshop aimed
to raise awareness on CountrySTAT, including its opportunities and challenges, based on the
Philippine experience and to provide discussions about metadata documents on national
agricultural statistical systems prepared by participating countries. The workshop also served
as the official launching of the CountrySTAT Philippines. The major output of the workshop was
the enhancement and streamlining of the FAO prescribed metadata outline that is agreed upon
and understood by participating countries. The Philippines has modified its metadata based on
the final FAO prescription. This was published last July 2007 and was made accessible via the
metadata link in the CountrySTAT Philippines webpage.
3.4 BAS In-house PC-Axis Software Training
The administration and updating of data and metadata in the CountrySTAT system suggested
additional work and time on the part of the Website operating unit of the BAS. Also, the PC-Axis
has evolved as a standard for disseminating statistical tables. These required more
involvement in terms of personnel who should be technically knowledgeable in using the PCAxis software. In solving this, the BAS conducted in-house PC-Axis software trainings in two
(2) batches. Technical operating units of the BAS were properly represented in the trainings.
The first batch was conducted in 27-29 November 2006, while the 2nd batch was conducted in
6-7 June 2007. The participants of the trainings will help in the conversion of data file to PX
files. The BAS plans to conduct similar trainings in the Provincial Operations Centers (POCs)
across the country.
8
4. Installation of the
CountrySTAT System



4.1. Installation Requirements
Just like in any other system, CountrySTAT has system requirements for its successful
installation. The CountrySTAT system is a Microsoft Windows-based application. This
should be installed in a Windows server either 2000 or 2003 with Internet Information
Server (IIS) version 5.0 or 6.0. The system requires that a PX-Web application with
license from Statistics Sweden be installed in the web server. The PX-Web is written in
ASP 3.0 and uses DLLs to handle all its functions. The directory where the
CountrySTAT files are located should be configured with READ and WRITE
authorization for Internet User Accounts (or IUSR), Launch Process Account (or IWAM),
and local group account (or IIS_WPG). The READ and WRITE authorization is
necessary since the system frequently creates temporary files to speed up the process
of presenting its output to the users. The functions that require temporary files are (1)
when a Query is made to the PX file, (2) when generating charts and maps, and (3)
when users download the generated report in other formats such as Excel or PC-Axis
file.
The Food and Agriculture Organization (FAO) suggested that the CountrySTAT system
should be installed as part of the website of the implementing agency. The
implementing agency should designate a CountrySTAT website administrator who is
knowledgeable in Active Server Pages (ASP) and Visual Basic Scripting (VBScript). The
designated administrator should also attend the CountrySTAT administration and PCAxis Software training which are provided by the FAO.
9
4. Installation of the
CountrySTAT System




4.2. Obtaining PC-Axis License
The NSCB had acquired a license for disseminating PC-Axis files in their previous project
with Statistics Sweden. Through email communications, the BAS contacted Statistics
Sweden and enquired on how to obtain a license for BAS similar to NSCB. Based on the
discussion with Statistics Sweden regarding the licensing terms on PC-Axis, a country
needs only one license. The agency who owns the license should report and request
permission from Statistics Sweden for another agency to be covered in the license.
Recognizing the terms, the NSCB required a separate MOA with the BAS on the exclusive
use of PC-axis in the CountrySTAT system. The MOA was signed on the 21 September
2006.
4.3. Enhancement of the CountrySTAT System
Prior to the installation of the CountrySTAT system, the designated CountrySTAT
Administrator did some modifications on the programs of the original CountrySTAT
system. The changes in the programs were due to the following: (1) installation of the
Philippines shape map; (2) making the system into a cross-browser; (3) enhancing and
speeding up the viewing of metadata in the Query interface, including the quick tables
and maps; and (4) linking of data tables with their corresponding metadata.
10
4. Installation of the
CountrySTAT System




4.4. Shift in Website Hosting Service
Until the early part of May 2006, the BAS website (http://www.bas.gov.ph) was hosted under a
shared server of Filipino Web Services, Inc. (FilWeb). The shared server operates in Linux
platform (Unix-based operating system) with Apache as its web server. The BAS website uses
Hypertext Pre-processor (PHP), PERL, and MySQL Database. All of these languages and
software are cross-platform which means that they can be installed either in MS Windows or
Unix-based (such as Linux) platforms.
In installing the CountrySTAT system, BAS made some adjustments in its website setup. The
FilWeb’s shared server cannot accommodate the CountrySTAT system since it requires
Windows platform and installation of PX-Web. FilWeb thought that the system might
compromise the operations of other websites in their shared server. As a result, the BAS
transferred its website to a dedicated server. The BAS website is now currently hosted by
ServoBox, a hosting service trademark of WebFocus Solutions, Incorporated. The dedicated
server operates using Windows 2003 server configured with IIS. It provides more benefits than
FilWeb’s shared server. The benefits would include faster accessibility, 99 percent uptime, no
set-up fee, affordable, less server administration on the part of the BAS. The server is installed
with updated Antivirus / Spam filtering, and more importantly only the BAS is using the server.
Transferring the website did not create major problems since the BAS website has been
developed using Cross-platform languages and software. The setting up and configuration of
the CountrySTAT system and the BAS website were done by the Systems Administrator of
ServoBox. The BAS Webmaster coordinated with ServoBox and provided the list of
requirements in order for the CountrySTAT system to work. The Philippine CountrySTAT system
as shown in figure 3 is working and has been made accessible since May 2006. It is a sub-
domain of the BAS website with URL http://countrystat.bas.gov.ph.
11
4. Installation of the CountrySTAT
System
 4.5. Initial Data
 In preparation for the first national training workshop, the
priority data sets to be loaded in the CountrySTAT were
already identified. The priority data were those that support
the statistical framework under the Core data set. The SRTC
which organized the workshop earlier advised the participants
to bring the required data with them in the training. The
training included exercises in converting their existing data
(in Excel format) into PX files. Towards the end of the training,
the resulting PX files were collected. The PX files were further
enhanced after finalizing the CountrySTAT guidelines. These
PX files served as the initial data of the CountrySTAT
Philippines during its installation in May 2006. .
12
5. The CountrySTAT Philippines




5.1. CountrySTAT Core Group
The CountrySTAT Core Group was created by the BAS management at the start of FAOPSA project. This was made up of selected personnel across operating units in the
Central Office. The group was tasked to do the following: (1) implement the CountrySTAT
activities and advocate for its use; (2) facilitate the consolidation of inputs to the
preparation of metadata report; and (3) serve as the Clearing House for the inputs to the
CountrySTAT. Thus, the continuous development and maintenance of the CountrySTAT
Philippines would largely rely on this group.
5.2. Major Domains
The CountrySTAT Philippines, as of July 2007, has provided access to 31,399,466 figures
(rows x columns) in 112 PX files (or statistical tables) distributed into 10 major domains.
The 10 major domains that comprise both the Core and Detail data sets for CountrySTAT
Philippines are the following: (1) Production, (2) Trade, (3) Food Consumption, (4) Prices,
(5) Agricultural Machinery, (6) Fertilizer, (7) Pesticides, (8) Land use, (9) Labor and
employment and (10) Others. The time series data in the Core data set cover at least 12
years. All the statistical tables posted in the CountrySTAT Philippines are adequately
backed up by metadata.
13
5. The CountrySTAT Philippines


5.3. File Naming Convention
The CountrySTAT Core Group developed and adopted a File naming
convention (see figure 4) in organizing the data in the CountrySTAT
Philippines webpage. This is included in the CountrySTAT guidelines. The
first two-character code represents the domain. This will be followed by a
three-character “statistics and indicators” code, and then followed by an
integer representing the agency code. The last two characters will serve as
the numerical id of the PX-file. The file name should be the same as the Matrix
keyword of the PX file. By just looking at the file name of the PX-file, one can
identify the location of the domain or category of the file and the source
agency. The linking of data to its metadata and vice versa is another
significant function of this convention.
14
4. Current Approaches in
Maintenance of Economic Accounts
for Food and Agriculture
5.4. Features and Outputs
The CountrySTAT Philippines is an applied system of PC-Axis. The
features or outputs of the CountrySTAT Philippines are similar to that
of the PC-Axis software. Figure 5 shows the sample generated table
from a query. Looking at the interface, one can easily change the
table format by transposition (or pivoting). Commonly used charts
can easily be generated to support data analysis. Conversion to
other file formats such as Excel, HTML, CSV and others is
straightforward that can used for data exchange. Thematic maps
such as shown in figure 6 can be easily generated. All of these
features can be made from a single click of a button.


5.5. Website Maintenance
The process of updating the CountrySTAT Philippines is illustrated in figure 7. Data
updates from other agencies such as NSO, BLES and NSCB are submitted via email by
their designated point persons. In the BAS, data updates are taken from the BAS
Centralized PC-Axis Database System (CPXDBS) by the designated database
coordinators and submitted thru the BAS Local Area Network (BASLAN). The CPXDBS is
a system developed by the BAS using PX-Web designed for walk in clients and
accessible only in the BASLAN. This system serves as a parallel system and source of
CountrySTAT updates. After receiving the updates, the CountrySTAT Core Group initiates
quality control and ensures that the PX-files are in conformity with the CountrySTAT
guidelines. If the PX files pass the quality check, then these will be submitted to website
operating unit of the BAS thru the BASLAN. The CountrySTAT administrator will upload 15
the PX-files to the server via FTP. If the uploads are completed successfully, the
4. Current Approaches in
Maintenance of Economic Accounts
for Food and Agriculture





Capital Formation Accounts
There has been an attempt to develop the capital formation accounts some years
ago. The lack of resources has prevented the BAS from implementing a survey on
agricultural capital formation which should have enabled the establishment of
benchmark data for the economic accounts system for agriculture. The framework
for the capital accounts identifies data requirements, defines concepts and describes
accounting principles and procedures. Basically, the capital account measures gross
fixed capital formation, changes in inventories and consumption of fixed capital and
net acquisition of lands by agricultural households.
Fixed capital assets include the following: land improvements and other construction
(work completed), plantation development, farm buildings, transport equipment and
machineries, tools and implements and work animals for breeding and milking and
layers. Stocks include the following: land improvement and plantation development
(work in progress), agricultural supplies and materials, harvested crops and
livestock for fattening and broilers.
About a decade ago, the then newly created BAS did a pilot capital formation survey
in a few provinces. The results of the survey became the basis for the
compilation of capital formation accounts. It was unfortunate, however, that for
financial reasons, the intended nationwide survey on capital formation in
agriculture was not pursued.

16
4. Current Approaches in
Maintenance of Economic Accounts
for Food and Agriculture

Capital Formation Accounts

In the conduct of the pilot test exercises re: capital formation, the BAS
was guided by the EAA Handbook. It followed the PSNA provisions for
the computation of capital formation for each of the item in the
account in which the value corresponding to the net acquisition of
capital items is being considered.
The following general concepts and procedures were used in the
preparation of capital formation accounts :
Gross fixed capital formation was taken as the difference between
producer’s acquisitions and disposals of new or existing fixed assets
during a given accounting period. This was recorded when ownership of
fixed assets was transferred to the institutional unit that would use
these assets in production. Similarly, assets produced for own account
were considered as gross fixed capital formation even as they were
being produced and were not completed or were not yet mature.
Fixed assets were valued at purchaser price so that valuation included
all costs involved in the transfer of ownership. In the case of
inventories, changes were reflected by subtracting the value of
inventories disposed of in the accounting period from the value of



inventories acquired by the enterprise. . TABLE
17
4. Current Approaches in
Maintenance of Economic Accounts
for Food and Agriculture



Agricultural Indicators System
The accounts discussed thus far served as tools to regularly monitor the
performance of the agriculture sector. To complement this objective, an
agricultural indicator system (AIS) has been conceptualized to help assess
the impacts of various policy measures on agriculture development and the
economic conditions of the population dependent on the sector.
The framework for the agricultural indicators system has been developed
with the intention of providing measures for assessing socio–
economic changes in the agriculture sector, characterizing the agrarian
structure of the economy and situating agriculture in the national
economy. In the process, the system has to consider the perspectives of
the development plans at the national, sectoral and sub-sectoral levels.

18
4. Current Approaches in
Maintenance of Economic Accounts
for Food and Agriculture
Agricultural Indicators System
The indicators system contains 16 broad categories namely:
 population and labor force
 economic growth
 poverty alleviation with equity
 agricultural structure and resources
 output and productivity
 inputs
 agricultural credit
 prices and marketing
 access to technical information
 exports and imports
 food consumption and nutrition
 food self – sufficiency and security
 access to community services
 redistribution of land
 people’s participation
 role of women in agriculture
19
5. Databases in Support of the
SEAFA





The compilation of various accounts under the SEAFA is dependent on the availability of
data. As far as the major data requirements of the production accounts and the supply
and utilization accounts are concerned, these are being supported by the current data
systems in agriculture. Please refer to the matrices on concepts and databases for
production and supply and utilization accounts.
Data Requirements of the Production Accounts : The data system on agricultural
production has to attune itself to the needs of agricultural and national accounting.
Regular surveys and monitoring activities should be conducted to generate production
and price data in the desired frequency, level of disaggregation and time lag.
For the “uses “ side of the production accounts, we need data on intermediate inputs.
These are obtained from costs of production surveys. As mentioned earlier, in the
Philippines, the task of estimating the gross value added in agriculture is taken up by
the central coordinating agency, the National Statistical Coordination Board ( NSCB ). To
date, the BAS has established a sizeable database on production costs which should
help in completing the data requirements for intermediate consumption.
An integral item in any of cost of production survey is the disposition of the produce.
This data set will allow delineation of output for the market and output for own final
use.
Another component of the production accounts refers to agricultural services. Data on
agricultural services are hardly available except for the relatively big-ticket items such as
irrigation, especially if this is run by the government. In this case, data can be obtained
from the irrigation department of the government.

20
5. Databases in Support of the
SEAFA

Data Requirements of Supply and Utilization Accounts : The data on
production are compiled on an annual basis. Data on imports and exports
are from the foreign trade statistics of the National Statistics Office. A major
area for improvement is the updating of parameters used in allocating
production to other uses such as processing. It is important that surveys or
special studies on utilization be done in order to update the utilization
parameters.

Data Requirements of Other Recommended Accounts : The data items indicated in
the generation of income account and the allocation of primary income account
for agricultural households can be generated by surveys on costs of production or
surveys on improved farm management. These surveys may also be able to
generate data inputs for compiling the capital formation account for the
households.
In the case of environmental accounts, the basic data requirements can be
supported by the agricultural data systems in two ways. One, some data items
may be collected through existing survey and monitoring activities. Two, a special
survey to establish benchmark data may be undertaken. The benchmark data can be
updated on a frequency established based on common requirements.

21
6. Advantages of the SEAFA


The 1996 SEAFA requires more data about agricultural households. Consequently it
offers much more information as inputs to policy formulation. The linkages of
transactions within the accounting framework make it easier for data users to
understand the agriculture sector in a broader context. The framework enables the
reader to relate production, income and capital accumulation by the agricultural
institutional units, with the households being the main unit, and also look at supply
and uses of agricultural goods and services produced by the sector. Specifically,
the SEAFA–recommended accounts allow international comparability using indicators
to monitor the status or progress of the sector. The various databases that will be
created for the SEAFA will require analysis of these data to produce indicators for
a structural analysis of the agricultural economy. Because the SEAFA relates in
terms of value aggregates, the system becomes the basis of quantity and price
indices. Valuation is done at both constant and current prices. Constant price valuation
enables aggregation of the outputs and inputs while taking into account differences in
quality, importance, location and other considerations.
The fact that the SEAFA requires comprehensive databases opens a big opportunity
for doing further analysis, the outputs of which become important inputs to policy.
In particular, the SEAFA databases can be used to support in-depth statistical analyses
and econometric modeling.

22
7. Implications of the 1993 SEAFA on
the Agricultural Statistical System







On Data Requirements
The 1996 SEAFA has definitely created additional pressure on agricultural statistical systems. In the Philippines, the
system has yet to fully satisfy the requirements of the 1974 EAA and already, it has to gear itself to address the
requirements of the 1996 SEAFA.
The sectoring of the accounts would require the agricultural data system to generate data for households as
distinguished from establishments. Presently, however, production data do not distinguish households from
establishments. This creates a problem since the household is the central institutional unit of interest in agriculture. The
recommended accounts or tables to be compiled within the 1996 SEAFA framework indicate the need for statistical
activities that can generate more detailed data to characterize agricultural households. The agricultural statistical
system has to reorient its commodity – focused data system.
Assuming that a statistical system has succeeded in developing and operating the agricultural accounts systems, its
main concern would then be the updating of data bases and parameters to improve the utility of these accounts. One
case in point is the need to update the parameters used to estimate utilization components like the share of
processing to total production or supply, the share of wastage, and the like.
Many national agricultural statistical services in the region have yet to come up with benchmark data on capital
formation in agriculture. The SNA has not had the benefit of obtaining updated benchmark data on capital
accumulation. With the adoption of the 1996 SEAFA, it has become all the more important to generate capital
formation data. Lastly, there is a need to do profiling of agricultural households to come up with benchmarks on
incomes and their sources and other household statistics to support the preparation of household sector accounts.
On Resource Requirements
As indicated earlier, the 1996 SEAFA requires databases that are comprehensive in the sense that they show different levels
or types of disaggregation and that they are available at desired frequency and schedule. All these characteristics mean
additional resources to support the conduct of necessary statistical activities to generate the data requirements of
the accounts and indicators systems. Country experiences show that government allotment for agricultural statistics
generally do not meet the financial requirements of the necessary statistical operations. Current resources are often just
enough to support the requirements for the conduct of production and price surveys.
23
8. Improving the Agricultural Service









The improvement of the agricultural statistical service is the underlying objective in developing an agricultural statistical
programme. It is recognised, however, that limited financial provisions for the operations of the statistical systems
have to be accepted as a given in this process.. The challenge to the statistical systems, therefore, has to do with
addressing the statistical needs of policy–makers and other data analysts and users given the inadequate
financial resources. In this respect, the use of statistical frameworks become highly relevant and critical.
A statistical framework is an integral part of statistical development planning. It guides the planning process and
facilitates the specification of desired results. A good appreciation of the frameworks will help in setting up priorities
of statistical activities viz a viz available resources. A framework spells out the current status of data systems
and therefore can guide in setting the goals/objectives of a desirable statistical system.
By itself, the SEAFA is important to policymakers as it provides information inputs to decision making. For the national
statistical system, however, the accounting framework further serves as the integrating mechanism for agriculture –
related statistics. As such, it defines the priority data that the agricultural statistical system has to consider in its
development planning. With the SEAFA as the background, statistical planning becomes focused and clear on the
following issues :
setting priorities in the use of available resources;
data support to development plans and programs;
data availability and gaps;
alternatives and innovative approaches in data generation and dissemination and others.
In more specific terms, the adoption of the SEAFA provisions calls for efforts to strengthen agricultural census
and survey programmes. In this respect, revisit of agricultural sampling frames and survey designs would be in
order. On the other hand, the system has to study the utility of data coming from administrative forms being
employed by many agencies.
To support the compilation of institutional accounts, the agricultural statistical system needs to generate/establish
databases by type of institutional unit particularly, the households. This can be addressed by means of an
improved farm household survey. The possibility of integrating or separating the conduct of costs of production
surveys and capital formation surveys must be studied. This survey can also support the data requirements for
the completion of both the uses and resources sides of the accounts.
24
8. Improving the Agricultural Service





1.
2.
Improving the agricultural data system will also mean looking into the updating requirements of basic parameters
used in accounting. These include conversion ratios for paddy rice, livestock meat and processing ratio in the
case of supply and utilization accounts. This would also be true for the source of nutrient equivalents, the FCT.
Apparently, some special surveys or studies are necessary.
The price data system should be able to support the valuation requirements of the accounts. Related to this is
the question about the relevance of the base year being used. Harmonizing base year for index computations
has to be considered for optimum use of indicators. Another consideration is the basket of commodities in the
price surveys. Is this being updated to suit the current conditions in the production and market spheres?
The SEAFA highlights the need to establish costs of production data bases. The agricultural statistical system
has to establish benchmark data on costs of production. This can be done by means of a sample survey, record
keeping, interview of key informants and any other means of data generation. Commodity – based studies
produce individual data by commodity while integrated surveys produce total farm cost data.
Balancing production and consumption statistics is a difficult but challenging task. In the absence of any recent
consumption survey, the supply and utilization accounts should help in rationalizing production and consumption
data. In this kind of analysis, the data reviewer or analyst must be armed with a thorough knowledge of the
available statistics and methodology used in generating these statistics so that he could explain the strengths and
weaknesses of his results.
SEAFA preparation should take into account the treatment of stocks and flows data in terms of valuation and
aggregation noting that stocks are results of accumulation of flows in the previous accounting periods and
these are changed by flows within the accounting period. Make better use of existing data; improve data
dissemination and use
Prepare master plans in support of PRSPs and other priority needs; involve producers and users (incl. decisionmakers)
Prioritize and define work for different budget levels and define level and modalities of funding from government budget
25
9. Conclusions





Even as some agricultural statistical systems are still polishing their
economic accounting systems, the SEAFA has come to redirect
integration in the statistical system. The development and
maintenance of agricultural statistical frameworks borne out of the
SEAFA require the agricultural statistical system to reassess its
current data systems and subsystems. As possible offshoots of this
revisit, the statistical system will address these concerns :
Which of the recommended accounts or tables are needed the most?
How can the recommended SEAFA tables be modified to suit policy
needs and match them with available data ?
What data are needed to generate SEAFA tables?
The adoption of the SEAFA by countries is desired but this process
cannot take place by simply conducting additional statistical activities
to produce additional data. The process has to be rationalised based on
the correct appreciation and understanding of SEAFA by statistical
designers and planners.
26
5. Implications for Donors
1. Support the preparation of country-owned master
plans in support of PRSP monitoring and
evaluation (incl. agricultural and rural subsystems)
2. Financially support Master Plans as a whole over
the medium to long term
3. Do not finance ad hoc surveys that are not part of
the Master Plan
4. Require local co-financing
5. Find balance between being pro-active catalysts
and respecting country ownership
27
5. Implications for Donors
(continued)
6. While supporting National Agr. Statistical System,
help to break down “silo mentality” and insure
integration with the national system
7. Support and/or undertake serious multi-country
research for estimating agricultural production in
general and for livestock, root crops, and
horticulture in particular
8. Support regional and sub-regional knowledge
training, knowledge sharing, etc.
28