71point4’s

Insights

Ideas, analysis and things we find worth exploring.

Insights · page 2

8–15 of 17
Policy & regulation

On the threshold

Inflation and statutory thresholds

Original thresholds compared with their end-2025 CPI-adjusted values.

Death
1 / 7
Original thresholdInflation-adjusted
Small estatesR250 000R429 883Magistrate reportingR125 000R273 936Estate duty abatementR3 500 000R9 196 429200020052010201520202025
Browse the seven categories
Death

Small estates: 2014-11, R250 000; end 2025, R429 883.

Magistrate reporting: 2009-07, R125 000; end 2025, R273 936.

Estate duty abatement: 2007-07, R3 500 000; end 2025, R9 196 429.

Taxes

Small business corporation tax: 2013-04, R20 000 000; end 2025, R37 184 116.

Turnover tax: 2009-03, R1 000 000; end 2025, R2 244 009.

VAT registration: 2009-03, R1 000 000; end 2025, R2 244 009.

Employment

Skills Development Levy: 2005-08, R500 000; end 2025, R1 479 885.

BEE exempt micro enterprise: 2015-05, R10 000 000; end 2025, R16 857 610.

Borrowing & lending

Short-term credit: 2007-06, R8 000; end 2025, R21 237.

Small juristic entity: 2006-06, R1 000 000; end 2025, R2 837 466.

Large credit agreement: 2007-06, R250 000; end 2025, R663 660.

Housing

RDP housing subsidy: 1995-11, R3 500; end 2025, R18 874.

FLISP: 2018-07, R22 000; end 2025, R31 429.

Social housing: 2022-07, R22 000; end 2025, R24 956.

Studying

NSFAS: 2017-12, R350 000; end 2025, R516 476.

Moving money

Foreign investment allowance: 2015-04, R10 000 000; end 2025, R16 940 789.

Single discretionary allowance: 2011-10, R1 000 000; end 2025, R2 007 797.

Source: SARB CPI · adjustment to December 2025

Frozen thresholds from the past reshape taxes, benefits, and markets. Inflation erodes intent, and blurs policy focus and yet hardly any laws have changed.

Illana Melzer
Read insight
Employment

The skills we never got

SETA training outcomes

Unemployed participants · 2012/13–2023/24

Enrolled in SETA programmes655 578100% of initial enrolments

Enrolments: 655578. Completions: 430535. Finding employment: 234641. Same baseline of 655578 in every stage.

Each figure = 1% of enrolments. Employment estimated from tracer studies. Source: the article’s SETA pipeline.

South Africa’s most expensive unemployment scheme

David Metelerkamp
Read insight
Employment

EE Targets Firm Level

Firms meeting employment equity targets

Private firms with 50+ employees · reported in 2023

Construction · 1 038 firms

Target metNot met
OccupationWomenMenWomen
& men
Top management
38%
42%
15%
Senior management
37%
40%
10%
Professional / middle
26%
48%
7%
Skilled technical
9%
75%
1%
Construction

1038 firms.

Top management: women 38%; men 42%; women and men 15% meet their targets.

Senior management: women 37%; men 40%; women and men 10% meet their targets.

Professional / middle: women 26%; men 48%; women and men 7% meet their targets.

Skilled technical: women 9%; men 75%; women and men 1% meet their targets.

Manufacturing

3050 firms.

Top management: women 34%; men 30%; women and men 11% meet their targets.

Senior management: women 35%; men 37%; women and men 8% meet their targets.

Professional / middle: women 37%; men 34%; women and men 5% meet their targets.

Skilled technical: women 31%; men 53%; women and men 5% meet their targets.

Finance & insurance

511 firms.

Top management: women 37%; men 30%; women and men 9% meet their targets.

Senior management: women 51%; men 30%; women and men 9% meet their targets.

Professional / middle: women 64%; men 18%; women and men 5% meet their targets.

Skilled technical: women 80%; men 12%; women and men 2% meet their targets.

Wholesale & retail

2094 firms.

Top management: women 33%; men 28%; women and men 8% meet their targets.

Senior management: women 41%; men 26%; women and men 5% meet their targets.

Professional / middle: women 47%; men 31%; women and men 6% meet their targets.

Skilled technical: women 52%; men 32%; women and men 3% meet their targets.

Source: Employment Equity Commission · 2023 private-sector submissions

Private sector transformation remains slow, with white South Africans still dominating top roles. Only two firms meet new equity targets – will stricter regulations drive change or damage the economy?

Illana Melzer
Read insight
Gambling

From Margins to Masses: Why Gambling’s Growth Demands a Regulatory Rethink

South African gambling turnover

Total money wagered · R billions · 2017/18–2023/24

Turnover
030060090012002017/182019/202021/222023/24
South African gambling turnover · Total money wagered · R billions · 2017/18–2023/24
DateTurnover
2017/18390
2018/19426
2019/20452
2020/21326
2021/22560
2022/23815
2023/241143
Source: National Gambling Board · published chart labels, rounded R billions

Gambling in South Africa has surged in recent years. This blog examines the changing dynamics of the gambling sector and offers potential regulatory solutions

Ryan Berelowitz
Read insight
Financial inclusion

The Closing of the Mobile Money Gender Gap in Rwanda

Rwanda · First-time users

How first-time users transact a year later

Transaction diversity, February–April 2024, for the April 2023 cohort.

Men under 35
Women under 35
Men 35 and over
Women 35 and over
No activityNo paymentsLow · 1–2 typesMedium · 3–4 typesHigh · 5+ types
Men under 3532% high payment diversityInitiated five or more types of digital transactions.
Transaction diversity, February–April 2024, by gender and age
SegmentNo activityNo digital paymentsLow payment diversityMedium payment diversityHigh payment diversity
Men under 3513%6%21%28%32%
Women under 3515%8%28%26%22%
Men 35 and over27%13%32%17%11%
Women 35 and over25%15%37%17%7%
Source: Mobile money transactions data, National Bank of Rwanda. Cohort: 59,500 first-time users.Digital transactions exclude cash-outs. Published percentages are rounded and may not sum to 100%.

Using supply-side data to monitor the uptake and usage of mobile money.

Claire Hayworth
Read insight
Data analytics

Why Simplicity Matters: Would You Rather Be Roughly Right or Precisely Wrong?

Median house value predictions in Boston neighbourhoods

ActualSimple analysisMachine learning models
Median house value · US$ thousands · scale shown per rowNeighbourhood 1202122actualNeighbourhood 2101520actualNeighbourhood 3283338actual
Published model predictions in US$ thousands
ModelNeighbourhood 1Neighbourhood 2Neighbourhood 3
Machine learning model 121.415.733.1
Machine learning model 221.920.929.6
Machine learning model 320.919.329.4
Simple analysis20.612.835.4
True value2111.836.2

Analytical sophistication and practical reliability are not the same thing. When faced with a data problem, it’s tempting to reach for the most technical, impressive-sounding method available. However, in many cases, a simpler analysis can produce similar results within a more robust framework.

Pieter Massyn
Read insight
Sport

No Contest: Fans Lose in SARU’s Pricing Game

Can South Africans afford rugby tickets?

Households by annual household income category*

R83 000 +: 754 993 households (4%)R83 000 +4%0.75mR50 000 – R83 000: 774 434 households (4%)R50 000 – R83 0004%0.77mR29 000 – R50 000: 1 395 719 households (7%)R29 000 – R50 0007%1.40mR8 300 – R29 000: 6 804 806 households (32%)R8 300 – R29 00032%6.80mR0 – R8 300: 11 552 872 households (54%)R0 – R8 30054%11.55m
R0 – R8 30011 552 872 households
54%of all households
Source: Stats SA · Income & Expenditure Survey 2022/23*Income is reported income less imputed rent.

The Springboks have never been more popular, but SARU’s ticket prices have put live Test rugby beyond the reach of most South Africans.

Ryan Berelowitz
Read insight
Data engineering

The data ended up in ClickHouse anyways

Kibana and Superset data pipelines

Kibana and Superset data pipelinesEmployment Equity forms are fetched, parsed and cleaned, then indexed in Elasticsearch as documents for Kibana. Transaction API data is fetched and cleaned, transformed into analytical tables in ClickHouse, then visualised in Superset.Employment Equity formsFetch · parse · cleanElasticsearchIndexed documents0KibanaTransaction APIFetch · cleanTransformClickHouseAnalytical tableSuperset

Two projects, two dashboards, different stacks – one in Kibana and one in Apache Superset, which came out on top?

David Metelerkamp
Read insight