This document reviews what procurement indicators can be calcultated using data from the United Kingdom, Kenya, Nepal, Uganda, South Africa and Zambia, as part of the OC4H project.

Coverage of key fields

The following table shows the coverage of key variables needed to calculate key procurement indicators:

Coverage of key fields
field GB KE NP UG ZA ZM
award_periodEndDate 0 0 0.6 6.2 0 48.9
award_periodStartDate 0 0 11.1 6.2 0 99.8
award_value_amount 37.9 13.2 0 18.8 3.6 49.1
award_value_currency 37.9 13.2 0 18.8 3.6 49.1
numberOfTenderers 0 0 100 0 0 99.4
ocid 100 100 100 100 100 100
procurementMethod 68.8 100 100 100 100 99.8
procuringEntity 100 100 100 100 9.3 100
supplier_name 54.1 13.2 0 10.4 100 49.1
tender_item_description 0 0 0 100 0 99.8
tender_periodEndDate 76.1 100 100 5.5 100 99.8
tender_periodStartDate 0 100 100 8.6 100 99.8
tender_status 100 100 100 100 100 99.8
tender_value_amount 16.7 0 100 100 0 99.8
tender_value_currency 16.7 0 100 100 0 99.8

Indicators that can be calculated

Based on the coverage of key fields, the following table shows the proportion of tenders that have the fields necessary to calculate key indictaros. For instance while the proportion of open tenders and cancelled tenders can be calculated for all the countries with a good coverage, other indicators such as the duration of the award period can be calculated only for 6% of the tenders in Uganda and 49% of the tenders in Zambia.

The full list of indicatos with their description can be found here

Coverage of key indicators
indicator GB KE NP UG ZA ZM
Average and median number of bidders per tender 0 0 100 0 0 99.4
Average duration of the award period 0 0 0.6 6.2 0 48.9
Average duration of the tendering period 0 100 100 5.3 100 99.8
Market concentration 0 0 0 10.4 0 49.1
Number of unique suppliers by institution 54.1 13.2 0 10.4 9.3 49.1
Proportion of cancelled tenders 100 100 100 100 100 99.8
Proportion of contracts with savings and overruns 7.2 0 0 18.8 0 49.1
Proportion of open tenders 68.8 100 100 100 100 99.8
Proportion of single bid tenders 0 0 100 0 0 99.4
Proportion of value awarded in single bid tenders vs competitive tenders 0 0 0 0 0 49.1
Share of direct awards 43.5 13.2 0 10.4 100 49.1
Share of single bid contracts 0 0 0 0 0 49.1
Top suppliers 37.7 13.2 0 10.4 3.6 49.1

Examples

Below are some examples of indicators calculated with the data.

Proportion of open tenders

Description: This indicator calculates the proportion of competitive tenders in the whole procurement market.

Considerations: Consider open tenders and other types of procurement methods that allow competition. It is important to check local regulations regarding the different types of procedures that apply and the threshold values for competitive tenders. This can be calculated for each procurement agency and by year.

Formula: (Number of open tenders / Total number of tenders)*100

Data fields needed: tender.procurementMethod

Interpretation: A higher value can signal more competition and integrity (transparency) while having more direct awards may signal a risk.

This indicator can be calculated for all the countries, however for GB 31% of the tenders do not have information about the procurement method used

direct limited open selective NA
GB 1.4 0.0 47.8 19.6 31.2
KE 0.0 0.0 100.0 0.0 0.0
NP 0.0 0.0 100.0 0.0 0.0
UG 93.2 6.8 0.0 0.0 0.0
ZA 0.0 0.0 100.0 0.0 0.0
ZM 0.0 83.0 16.8 0.0 0.2

Proportion of single bid tenders vs competitive tenders

Description: This indicator calculates the proportion of tenders that received a single bid out of the total number of tenders were competition was expected.

Considerations: This has to be calculated for open (or limited selected) tenders where competition is expected. Single bids can be analysed by procuring entity or by item category. For some markets (or items) single bids can be a result of specialised goods or limited suppliers.

Formula:(TOTAL Number tenders with a single bid/ TOTAL number of tenders)*100

Interpretation: Having a high proportion of single bids in tenders that should be competitive is considered a risk in procurement, since it might signal limited competition, a low capacity of procurement agencies, result in higher prices, lower quality, weaker political accountability.

Data fields needed: tender.procurementMethod, tender.numberofTenderers or tender.tenderers_count

The number of tenderers is only available for Zambia and Nepal

Proportion of value awarded in single bid tenders vs competitive tenders

Description: This indicator calculates the total value awarded in tenders that received a single bid in comparison to the value awarded in competitive tenders.

Considerations: This has to be calculated for open (or limited selected) tenders where competition is expected. Single bids can be analysed by procuring entity or by item category. For some markets (or items) single bids can be a result of specialised goods or limited suppliers.

Formula:(TOTAL Award value of tenders with a single bid/ TOTAL value awarded)*100

Interpretation: Having a higher value awarded in non competitive tenders may signal less competition in high value tenders, or higher prices as a result of no competition.

Data fields needed: tender.procurementMethod, tender.numberofTenderers or tender.tenderers_count, award.value

Only Zambia hast both the number of tenderers and the award values. In the case of Nepal they dont publish award values, and for thwe rest of the countries, the number of bidders is not available so it is not posible to segment between single bid and competitive tenders.

Flagged tenders

Based on the flags included in the dataset, related to constrained competition and health, all the tenders have at least one of the flags.

GB has the higher proportion of health tenders with 16.6%, while in other countries like Zambia only 4.7% of tenders where related to health. A high proportion of the tenders have at least one flag related to constrained competition. For instance in South Africa a 100% of the tenders were flagged, this could be an error in the flags definition. Similarly in Kenya and Zambia more than 80% of the tenders are flagged, so it could be worth verifying if the flag definition and thresholds are correct.

country both competition_constrained health
GB 5.8 37.3 16.6
KE 6.2 81.1 7.4
NP 2.3 28.5 8.5
UG 0.0 0.6 14.2
ZA 5.7 100.0 5.7
ZM 3.3 88.3 4.7