Food Balance Sheets FBS component Seed Learning Objectives

Food Balance Sheets FBS component: Seed

Learning Objectives At the end of this session, the audience will know: • Different data sources for seed. • Recommended approach for Imputation and estimation of seed use. 2

Outline • Definitions • Data sources • Imputation and Estimation 3 3

1 Definitions 4 4

Definition • Seed is defined as any quantity of a commodity set aside for reproductive purposes in the following year. This can include ⁻ seed for sowing, ⁻ plants for transplanting, ⁻ eggs for hatching, and ⁻ fish used as bait. • This quantity should also take into account double or successive sowing. 5 5

2 Data sources 6 6

Data sources • Most official measurements of seed use data are sourced from agricultural surveys. • Most surveys will include questions on both purchases of improved seed and quantities of own reserved seed, • What to do if the survey excludes purchases of improved seed? ⁻ Possible to access the sales records of commercial seed companies in order to obtain a full estimate of total seed use. ⁻ Trade data may also give some indications about seed quantities if most seed is imported, as seed typically have separate HS codes. 7 7

Data sources • What to do if an estimate of total seed use is not available and a historical seeding rate cannot be calculated from the data? ⁻ Investigate the possibility of whether or not there is available data regarding only seeding rates. ( technical conversion factor can be a useful source for that) ⁻ Information on either optimal or effective seeding rates may be available from seed companies or extension specialists, for example. ⁻ If the typical seed rate is known, then imputing total seed use is a simple calculation, according to the methodology described below 8 8

3 Imputation and estimation 9 9

Imputation and estimation • 10 10

Imputation and estimation • Given this identity, the process of deriving an imputed value for seed quantity is as follows: Step 1: • Calculate/estimate a seeding rate. Step 2: • If missing, impute a value for sown area in the following year. Step 3: • Multiply the two values together for an estimate of total seed use. 11 11

Imputation and estimation • 12 12

Imputation and estimation • Step 2: Area Imputation • If an estimate for sown area in the following year is available, then skip to Step 3 • If no estimate for sown area in the following year is available, then sown area must be imputed. • This can be done through one of three approaches, depending upon the data available to the country-level compiler. 13 13

Imputation and estimation • 14 14

Imputation and estimation • 15 15

Imputation and estimation • 16 16

FBS component: Seed Imputation and estimation • 17 17

Imputation and estimation • 18 18

FBS component: Seed Imputation and estimation • Step 3: Multiply the two values • This approach should only be used when either of the two options above are not possible. • Once a seeding rate and sown area in t+1 have been estimated for the product in question, the two values are multiplied to arrive at the quantity of seed needed in year t. 19 19

Imputation and estimation Exercise 1 • Year Sown area (HA) (A) Harvested area (HA) (B) 2010 400 388 2011 425 405 2012 420 395 2013 390 370 2014 ? 385 20 20

Imputation and estimation Correction Year Sown area (HA) (A) 400 Harvested area (HA) (B) 388 2010 400/388 = 1. 03 2011 425 405 425/405 = 1. 05 2012 420 395 420/395 = 1. 06 2013 390 370 390/370 = 1. 05 2014 404 385 21 21

Imputation and estimation Exercise 2 • 22 22

References 3 rd Chapter of the guideline: ‘Data for FBS compilation: considerations, sources and imputation’, 3. 5. 7 Seed 23 23

Thank You