Differentiate interest rates by SSP in scenario_config

[flohump]

🐦 Description of this PR 🐦

• Add SSP-specific interest rates (s12_interest_lic, s12_interest_hic) to scenario_config.csv. Previously, all SSPs used the same default rates (LIC=10%, HIC=4%), which is inconsistent with the SSP narratives.

SSP	LIC	HIC	Rationale
SSP1	10%	4%	unchanged
SSP2	15%	6%	moderate institutional barriers
SSP3	20%	8%	fragmentation, restricted capital flows
SSP4	20%	6%	highest inequality — LIC left behind, HIC relatively stable
SSP5	10%	4%	unchanged

• SDP and VLLO scenarios remain at SSP1 defaults.
• Historical rates are untouched; new rates phase in linearly between 2025 and 2050 via the existing fader mechanism.
• default.cfg updated to SSP2 values (LIC=15%, HIC=6%) since SSP2 is MAgPIE's default scenario. The noselect variants were updated accordingly.

Motivation

The interest rate is fundamental for the balance between land expansion and intensification in MAgPIE. It determines the cost of capital for agricultural investments. Low interest rates favor intensification; high rates shift the balance toward land expansion.

Previously, all SSPs used identical rates, implicitly assuming the same investment environment regardless of scenario narrative. This led to uniformly high intensification and crop yields across SSPs — inconsistent with the SSP framework and an outlier compared to other IAMs in ScenarioMIP.

The new rates are consistent with observed agricultural lending rates in Sub-Saharan Africa, where unsubsidized agribusiness loans average more than 20% (World Bank Group 2016), and with commercial lending rates in OECD countries of 3–7% (World Bank 2024, indicator FR.INR.LEND) https://data.worldbank.org/indicator/FR.INR.LEND.

Test plan

• Compare SSP2 crop yields and land-use change against previous default
• Verify SSP1 and SSP5 results are unchanged

Results

SSP2-NPI

Tables with global values. SSP1 and SSP5 are unchanged (as expected) and omitted from the table.

Cereal yields (t DM/ha)

Scenario	2050 develop	2050 thisPR	Δ%	2100 develop	2100 thisPR	Δ%
SSP2-NPi2025	4.31	4.20	-2.5%	4.49	4.20	-6.6%
SSP2-PkBudg1000	4.70	4.57	-2.8%	5.86	5.42	-7.4%
SSP2-PkBudg650	5.46	5.17	-5.3%	6.45	5.96	-7.7%
SSP3-NPi2025	4.36	3.98	-8.7%	4.54	3.69	-18.8%
SSP3-PkBudg1000	4.94	4.46	-9.6%	6.28	5.23	-16.7%
SSP3-rollBack	4.34	3.97	-8.6%	4.42	3.57	-19.3%
SSP4-NPi2025	4.35	4.24	-2.7%	4.59	4.10	-10.6%

Bioenergy crop yields (t DM/ha)

Scenario	2050 develop	2050 thisPR	Δ%	2100 develop	2100 thisPR	Δ%
SSP2-NPi2025	15.0	14.1	-6.5%	24.3	23.0	-5.3%
SSP2-PkBudg1000	18.9	17.7	-6.6%	24.6	23.3	-5.5%
SSP2-PkBudg650	20.0	19.3	-3.8%	23.6	22.9	-2.7%
SSP3-NPi2025	16.1	14.5	-10.3%	26.1	25.5	-2.3%
SSP3-PkBudg1000	20.4	18.6	-8.9%	25.0	22.5	-9.9%
SSP3-rollBack	13.7	11.8	-13.6%	27.1	26.0	-4.1%
SSP4-NPi2025	15.4	14.3	-7.2%	25.8	24.7	-4.0%

Land use intensity (Tau index)

Scenario	2050 develop	2050 thisPR	Δ%	2100 develop	2100 thisPR	Δ%
SSP2-NPi2025	1.64	1.55	-5.3%	1.88	1.75	-7.1%
SSP2-PkBudg1000	1.73	1.63	-6.0%	2.37	2.19	-7.9%
SSP2-PkBudg650	1.96	1.84	-6.2%	2.59	2.40	-7.4%
SSP3-NPi2025	1.69	1.50	-10.9%	1.95	1.61	-17.6%
SSP3-PkBudg1000	1.83	1.61	-12.1%	2.55	2.18	-14.6%
SSP3-rollBack	1.67	1.49	-11.1%	1.92	1.58	-17.7%
SSP4-NPi2025	1.65	1.53	-6.8%	1.91	1.70	-11.1%

Cropland (million ha)

Scenario	2050 develop	2050 thisPR	Δ%	2100 develop	2100 thisPR	Δ%
SSP2-NPi2025	1663	1695	+2.0%	1930	2042	+5.8%
SSP2-PkBudg1000	1573	1625	+3.3%	1691	1818	+7.5%
SSP2-PkBudg650	1582	1656	+4.7%	1674	1790	+6.9%
SSP3-NPi2025	1703	1850	+8.6%	2210	2650	+19.9%
SSP3-PkBudg1000	1597	1748	+9.5%	1876	2190	+16.7%
SSP3-rollBack	1705	1841	+8.0%	2238	2706	+20.9%
SSP4-NPi2025	1656	1707	+3.1%	2087	2295	+10.0%

LUC CO2 emissions (Mt CO2/yr)

Scenario	2050 develop	2050 thisPR	Δ	2100 develop	2100 thisPR	Δ
SSP2-NPi2025	182	385	+203	1286	1449	+163
SSP2-PkBudg1000	-2315	-1936	+379	-2846	-2129	+716
SSP2-PkBudg650	-3021	-2674	+348	-3001	-2393	+608
SSP3-NPi2025	1202	3719	+2517	2489	8173	+5684
SSP3-PkBudg1000	-1982	-1389	+593	-1782	-652	+1130
SSP3-rollBack	1548	3277	+1729	3513	7805	+4292
SSP4-NPi2025	210	203	-7	2797	5004	+2206

Key takeaways

1. SSP1 and SSP5 completely unchanged — confirming the marker scenario (SSP1-VLLO) is unaffected.
2. Yields decrease as intended: SSP2 cereal yields drop 3–8%, SSP3 drops 9–19% by 2100. The effect is stronger for SSP3, consistent with the larger rate increase.
3. Tau (land use intensity) decreases — this is the main driver behind the lower crop yields, together with changing cropland patterns.
4. Cropland expands to compensate for lower yields: +2–8% in SSP2, +8–21% in SSP3 by 2100. This is the expected intensification→expansion shift.
5. LUC CO2 emissions increase substantially, especially in SSP3 baseline scenarios where the cropland expansion into forests drives large additional emissions. In mitigation scenarios the effect is moderated by carbon pricing but still significant.
6. SSP3 effects are very large — SSP3-NPi2025 and SSP3-rollBack show cropland increases of ~440–470 Mha and LUC emissions increases of 4–6 Gt CO2/yr by 2100.

---

Data source: report_all.rds. Analysis created with Claude Code.

🔧 Checklist for PR creator 🔧

If a point is not applicable, check the checkbox anyway and write "non-applicable" next to the checkbox.

• Label pull request from the label list.

  • Low risk: Simple bugfixes (missing files, updated documentation, typos) or changes in start or output scripts
  • Medium risk: Uncritical changes in the model core (e.g. moderate modifications in non-default realizations)
  • High risk: Critical changes in model core or default settings (e.g. changing a model default or adjusting a core mechanic in the model)

• Self-review own code

  • No hard coded numbers and cluster/country/region names.
  • The new code doesn't contain declared but unused parameters or variables.
  • magpie4 R library has been updated accordingly and backwards compatible where necessary.
  • scenario_config.csv has been updated accordingly (important if default.cfg has been updated)

• Document changes

  • Add changes to CHANGELOG.md
  • Where relevant, put In-code documentation comments
  • Properly address updates in interfaces in the module documentations
  • run goxygen::goxygen() and verify the modified code is properly documented

• Perform test runs

  • Low risk:
    • Run a compilation check via Rscript start.R --> "compilation check"
  • Medium risk:
    • Run default run via Rscript start.R --> "default"
    • Check logs for errors/warnings
    • Fill in performance section below
  • High risk:
    • Run test runs via Rscript start.R --> "test runs"
    • Check logs for errors/warnings
    • Default run from the PR target branch for comparison
    • Provide relevant comparison plots (land-use, emissions, food prices, land-use intensity,...)
    • Fill in performance section below

• Reporting produces no errors and no new warnings

• Get two approving reviews (at least one from RSE)

📉 Performance 📈

• This PR's default : 28 mins
• Current develop default: 28 min

🚨 Checklist for reviewer 🚨

• PR is labeled correctly
• Code changes look reasonable
  • No hard coded numbers and cluster/country/region names.
  • No unnecessary increase in module interfaces
  • model behavior/performance is satisfactory.
• Changes are properly documented
  • CHANGELOG is updated correctly
  • Updates in interfaces have been properly addressed in the module documentations
  • In-code documentation looks appropriate

[pvjeetze]

Further discussions are ongoing between @flohump, @bodirsky and me