Full Text Available

Note: Clicking the button above will open the full text document at the original institutional repository in a new window.

The long-run returns to public agricultural R&D in South Africa: evidence from an updated TFP Series (1950–2023)

Thesis (MScAgric)--Stellenbosch University, 2026.

Saved in:
Bibliographic Details
Main Author: Molokoli, Poloko John
Other Authors: Greyling, Jan Christoffel
Format: Thesis
Language:English
Published: Stellenbosch : Stellenbosch University 2026
Subjects:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1869484297573892096
access_status_str Open Access
author Molokoli, Poloko John
author2 Greyling, Jan Christoffel
author_browse Greyling, Jan Christoffel
Molokoli, Poloko John
author_facet Greyling, Jan Christoffel
Molokoli, Poloko John
author_sort Molokoli, Poloko John
collection Thesis
dc_rights_str_mv Stellenbosch University
description Thesis (MScAgric)--Stellenbosch University, 2026.
format Thesis
id oai:scholar.sun.ac.za:10019.1/136158
institution Stellenbosch University (South Africa)
language English
last_indexed 2026-07-01T04:12:39.520Z
license_str Other — see source repository
provenance_str_mv Harvested via OAI-PMH from SUNScholar — Stellenbosch University Repository
publishDate 2026
publishDateRange 2026
publishDateSort 2026
publisher Stellenbosch : Stellenbosch University
publisherStr Stellenbosch : Stellenbosch University
record_format dspace
source_str SUNScholar — Stellenbosch University Repository
spelling oai:scholar.sun.ac.za:10019.1/136158 The long-run returns to public agricultural R&D in South Africa: evidence from an updated TFP Series (1950–2023) Molokoli, Poloko John Greyling, Jan Christoffel Stellenbosch University. Faculty of AgriSciences. Dept. of Agricultural Economics. Agricultural productivity -- South Africa Agriculture -- Research -- Economic aspects -- South Africa Total factor productivity -- South Africa Research and development projects -- Economic aspects -- Econometric models UCTD Thesis (MScAgric)--Stellenbosch University, 2026. Molokoli, P. J. 2026. The Long-Run Returns to Public Agricultural R&D in South Africa: Evidence from an Updated TFP Series (1950–2023). Unpublished masters thesis. Stellenbosch: Stellenbosch University [online]. Available: https://scholar.sun.ac.za/items/feec0c08-945b-4f1e-91cd-42282d697999 ENGLISH ABSTRACT: Sustained growth in South African agriculture increasingly depends on productivity improvements rather than continued expansion of land, labour, and other inputs. This thesis updates and extends South Africa’s long-run agricultural total factor productivity (TFP) series and quantifies the long-run productivity returns to public agricultural research and development (R&D). The study addresses two questions: (i) how has national agricultural TFP evolved since 1950 once the canonical series is extended beyond 2010; and (ii) what are the long-run productivity effects and economic returns to public agricultural R&D when research benefits are modelled as delayed and persistent and weather variability is controlled for? To update productivity measurement, the thesis extends the Liebenberg (2012) national series to 2023 using a Törnqvist index-number approach to aggregate heterogeneous outputs and inputs into annual output and input indices, with TFP calculated as their ratio and normalised to 2020 = 100. The updated series indicates three broad phases: relative stagnation from 1950 to the early 1970s, a sharp productivity rise from the mid-1970s to a peak in the late 1980s, and a subsequent decline followed by stabilisation around the 2020 benchmark from the mid-2000s through 2023. To estimate the R&D–productivity relationship, public agricultural R&D expenditure is converted into an R&D knowledge stock constructed as a weighted sum of past spending using alternative truncated gamma lag distributions with a two-year gestation period. The long-run association between log TFP and log knowledge stocks is estimated within a cointegrating Dynamic Ordinary Least Squares (DOLS) framework, controlling for weather variability using an Agricultural Weather Index derived from detrended national crop yields and allowing for alternative deterministic time controls (including a piecewise trend with a structural break in 1988). After econometric screening of 280 specifications, the best-performing models imply that the peak productivity effect of public R&D occurs roughly 11–14 years after expenditure. Estimated long-run elasticities of TFP with respect to the R&D knowledge stock range from 0.241 to 0.361 (median 0.299), and are robust across admissible lag specifications and trend controls. Elasticity estimates are translated into marginal benefit–cost ratios (BCRs) under alternative real discount rates. Across preferred specifications, BCRs remain well above one, ranging from 10.31 to 14.79 at a 3% discount rate and from 5.58 to 7.18 at an 8% discount rate. Overall, the results provide updated evidence that South African public agricultural R&D has generated substantial long-run productivity gains with delayed but persistent impacts, supporting the case for stable research investment as a central component of productivity-led and sustainable agricultural growth. AFRIKAANSE OPSOMMING: Volgehoue groei in die Suid-Afrikaanse landbou berus toenemend op produktiwiteitsverbeterings eerder as op die voortgesette uitbreiding van grond, arbeid en ander insette. Hierdie tesis werk Suid-Afrika se langtermyn landbou-totale faktorproduktiwiteit (TFP) reeks by en brei dit uit, en kwantifiseer die langtermyn produktiwiteitsopbrengste van openbare landbou-navorsing en -ontwikkeling (N&O). Die studie spreek twee vrae aan: (i) hoe het nasionale landbou-TFP sedert 1950 ontwikkel wanneer die kanonieke reeks verder as 2010 uitgebrei word; en (ii) wat is die langtermyn produktiwiteitseffekte en ekonomiese opbrengste van openbare landbou-N&O wanneer navorsingsvoordele as vertraag en volgehou gemodelleer word en weer-verwante wisselvalligheid beheer word? Om produktiwiteitsmeting by te werk, brei die tesis die Liebenberg (2012) nasionale reeks tot 2023 uit deur ’n Törnqvist-indeksbenadering te gebruik om heterogene uitsette en insette in jaarlikse uitset- en insetindekse te aggregeer, met TFP bereken as hul verhouding en genormaliseer tot 2020 = 100. Die bygewerkte reeks dui op drie breë fases: relatiewe stagnasie van 1950 tot die vroeë 1970’s, ’n skerp produktiwiteitsstyging vanaf die middel-1970’s tot ’n piek in die laat 1980’s, en daarna ’n afname gevolg deur stabilisering rondom die 2020-verwysingsvlak vanaf die middel-2000’s tot 2023. Om die N&O–produktiwiteit-verhouding te beraam, word openbare landbou-N&O-uitgawes omgeskakel na ’n N&O-kennisvoorraad wat as ’n geweegde som van vorige besteding saamgestel word met behulp van alternatiewe afgekorte gamma-vertragingsverdelings, met ’n tweejaar-gestasiefase. Die langtermynverband tussen die log van TFP en die log van kennisvoorrade word beraam binne ’n gekoïntegreerde Dinamiese Gewone Kleinste Kwadrate (DOLS) raamwerk, met beheer vir weer-verwante wisselvalligheid deur ’n Landbou-Weerindeks wat uit gedetrende nasionale gewasopbrengste afgelei is, en met alternatiewe deterministiese tydbeheer (insluitend ’n stuksgewys tendens met ’n strukturele breuk in 1988). Ná ekonometriese sifting van 280 spesifikasies impliseer die beste-presterende modelle dat die piek produktiwiteitseffek van openbare N&O ongeveer 11–14 jaar ná besteding voorkom. Beraamde langtermyn-elastisiteite van TFP met betrekking tot die N&O-kennisvoorraad wissel van 0.241 tot 0.361 (mediaan 0.299), en is robuust oor aanvaarbare vertragingspesifikasies en tendensbeheer. Elastisiteitsberamings word omgeskakel na marginale voordeel–koste-verhoudings (VKR’e) onder alternatiewe reële verdiskonteringskoerse. Oor die voorkeurspesifikasies bly VKR’e ruim bo een, met ’n reeks van 10.31 tot 14.79 teen ’n 3%-verdiskonteringskoers en van 5.58 tot 7.18 teen ’n 8%-verdiskonteringskoers. Oor die algemeen bied die resultate bygewerkte bewyse dat openbare landbou-N&O in Suid-Afrika beduidende langtermyn produktiwiteitswinste opgelewer het met vertraagde maar volgehoue impakte, en ondersteun dit die argument vir stabiele navorsingsbelegging as ’n kernkomponent van produktiwiteitsgedrewe en volhoubare landbougroei. Masters 2026-04-23T13:14:33Z 2026-04-23T13:14:33Z 2026-03 Thesis https://scholar.sun.ac.za/handle/10019.1/136158 en Stellenbosch University v, iii, 45 pages : illustrations application/pdf Stellenbosch : Stellenbosch University
spellingShingle Agricultural productivity -- South Africa
Agriculture -- Research -- Economic aspects -- South Africa
Total factor productivity -- South Africa
Research and development projects -- Economic aspects -- Econometric models
UCTD
Molokoli, Poloko John
The long-run returns to public agricultural R&D in South Africa: evidence from an updated TFP Series (1950–2023)
title The long-run returns to public agricultural R&D in South Africa: evidence from an updated TFP Series (1950–2023)
title_full The long-run returns to public agricultural R&D in South Africa: evidence from an updated TFP Series (1950–2023)
title_fullStr The long-run returns to public agricultural R&D in South Africa: evidence from an updated TFP Series (1950–2023)
title_full_unstemmed The long-run returns to public agricultural R&D in South Africa: evidence from an updated TFP Series (1950–2023)
title_short The long-run returns to public agricultural R&D in South Africa: evidence from an updated TFP Series (1950–2023)
title_sort long run returns to public agricultural r d in south africa evidence from an updated tfp series 1950 2023
topic Agricultural productivity -- South Africa
Agriculture -- Research -- Economic aspects -- South Africa
Total factor productivity -- South Africa
Research and development projects -- Economic aspects -- Econometric models
UCTD
url https://scholar.sun.ac.za/handle/10019.1/136158
work_keys_str_mv AT molokolipolokojohn thelongrunreturnstopublicagriculturalrdinsouthafricaevidencefromanupdatedtfpseries19502023
AT molokolipolokojohn longrunreturnstopublicagriculturalrdinsouthafricaevidencefromanupdatedtfpseries19502023