Add WACC module and a new realization in the MACRO module about investment inefficiencies by daymontas1 · Pull Request #2278 · remindmodel/remind

daymontas1 · 2026-01-29T10:58:04Z

Capital Market Modelling in REMIND.pptx
Cost_of_Capital_Documentation_v2.docx

Purpose of this PR

This PR introduces (1) a new realization in the macro module that accounts for investment losses arising from domestic capital market inefficiencies, and (2) a new module (Module 25_WACC) for calculating capital costs associated with energy-system and macroeconomic investments.

Description

Investment_Inefficiencies
The Investment_Inefficiencies realization in the macro module computes the share of new investments that is not transformed into productive capital due to financial frictions, misallocation, installation costs, or other inefficiencies. A parameter, φ, is introduced into the capital accumulation equation to represent these inefficiencies.

Specifically, φ captures the share of gross investment that successfully translates into productive capital in each region. Regional φ values are calibrated using literature-defined bounds* and empirical data on the efficiency of financial systems.

Module 25_WACC
Module 25_WACC computes, for each time period, technology-specific weighted average cost of capital (WACC)–related costs for the energy system, as well as costs associated with macroeconomic investment flows.

Energy-technology WACC costs consist of:
(i) costs associated with newly installed capacity in the current period, and
(ii) residual costs of technologies that remain in operation from previous periods.

This structure relies on the assumption that the repayment period equals the technical lifetime of each technology. Input WACC data are primarily drawn from Calcaterra et al. and are aggregated to REMIND regions using GDP-weighted averages of the countries comprising each region.

From these values, WACC mark-ups are derived and defined as the difference between the WACC of a given technology–region pair and the lowest observed WACC. The latter corresponds to natural-gas power generation in the USA and the CAZ region (Canada, Australia, and New Zealand), which serves as the risk-free reference case (i.e., zero mark-up) against which all other combinations are benchmarked.

To preserve budget neutrality, WACC-related costs are recycled back into the economy. At the same time, to capture the stimulating effect of WACC costs on investment decisions, WACC costs calculated in the previous model iteration are deducted from those computed in the current iteration. As the model converges, the net WACC cost component in the budget constraint approaches zero.

Further documentation is provided in the attached presentation and Word file.

Parts concerned

◻️ GAMS Code
◻️ R-scripts
◻️ Documentation (GAMS incode documentation, comments, tutorials)
◻️ Input data / CES parameters
◻️ Tests, CI/CD (continuous integration/deployment)
◻️ Configuration (switches in main.gms, default.cfg, and scenario_config*.csv files)
◻️ Other (please give a description)

Impact

◻️ Bug fix
◻️ Refactoring
◻️ New feature
◻️ Change of parameter values or input data (including CES parameters)
◻️ Minor change (default scenarios show only small differences)
◻️ Fundamental change of results of default scenarios

Checklist

Do not delete any line. Leave unfinished elements unchecked so others know how far along you are.
In the end all checkboxes must be ticked before you can merge.

I executed the automated model tests (make test) after my final commit and all tests pass (FAIL 0)
I adjusted the reporting in remind2 if and where it was needed
I adjusted the madrat packages (mrremind and other packages involved) for input data generation if and where it was needed
My code follows the coding etiquette
I explained my changes within the PR, particularly in hard-to-understand areas
I checked that the in-code documentation is up-to-date
I adjusted forbiddenColumnNames in readCheckScenarioConfig.R in case the PR leads to deprecated switches
I updated the CHANGELOG.md correctly (added, changed, fixed, removed, input data/calibration)

Further information (optional)

Runs with these changes are here:
Comparison of results (what changes by this PR?):

daymontas1 · 2026-01-30T15:56:41Z

PR in mrremind for the required input data here: pik-piam/mrremind#778

nicobauer

HI Diamantis. I left a few comments. Mostly they are on deletions of unnecessary lines that are commented anyways. ONly in the postsolve for the WACCS are two more important comments. Thanks and see you latter.

nicobauer · 2026-02-02T09:06:06Z

+*  WACC implementation on capital costs
+***---------------------------------------------------------------------------
+***This equation calculates the weighted average cost of capital (WACC) costs for new and existing technologies while ensuring that each past investment is assigned the correct WACC from the year
+q25_techwaccCost(t, regi)$(t.val > 2005)..


Do I see correctly that the tech-wacc parameters already include the country specific wacc parameters?

nicobauer · 2026-02-02T09:14:07Z

+  sum((t2)$((t.val - t2.val <= pm_lifetime_max(regi,tewacc)) and (t2.val < t.val)),  
+      (vm_costInvTeDir.l(t2, regi, tewacc)         
+       + vm_costInvTeAdj.l(t2, regi, tewacc)$teAdj(tewacc)  
+      ) * p25_techwacc(t2, regi, tewacc) * pm_ts(t2)/2);


The /2 is the critical parameter that you wanted to x-check with Ropert Pietzcker. Right?

daymontas1 · 2026-03-26T11:06:27Z

@LaviniaBaumstark Hi Lavinia, before finalizing this PR, there is an issue we need to address. With the capital markets implementation, I face convergence issues in REMIND. I have mainly addressed these by decreasing the lifetime of macro investments.

In the previous REMIND version, I increased the limit used in the respective equation (see slide 2 in the attached presentation to this PR, equation investment lifetime) from 1% to 10%, and the model converged
(p/tmp/adamanti/remind_imperfect_markets_b/output/SSP2-NPi2025_2025-12-04_17.16.10). In the latest version, 10% did not work (p/tmp/adamanti/remind_imperfect_markets_b/output/SSP2-NPi2025_imperfectb_2026-02-20_11.40.31),
and I had to increase it even further. For example, with 50% it converged:
(p/tmp/adamanti/remind_imperfect_markets_b/output/SSP2-NPi2025_imperfectb_2026-03-19_00.01.37).

Is it possible to relax REMIND in some way so that we deal with this more fundamentally while keeping the investment lifetime at its originally planned level (in principle, 1% of the original value)? Or close to that level ? Thanks!

LaviniaBaumstark · 2026-03-30T11:22:33Z

On Friday there was a (quick)fix for non-converging PkBudg-scenarios. I would try this. If possible, pull from develop branch

…tment inefficiencies

daymontas1 · 2026-06-15T15:35:58Z

PR in mrremind for the required input data here: pik-piam/mrremind#778

The mrremind PR with the input data of this REMIND PR was merged: pik-piam/mrremind#822

…1/remind into imperfect-markets-wacc

nicobauer · 2026-06-18T13:30:38Z

I checked the calibration file for the macro-parameters. I find several gaps in the red-labeled calibrations (not for SSA, which was our bad case, but other countries). For example in the US I find this gap for late periods. We can see that the scenario wants the trajectory to decrease, but the model seems unable. This makes me wonder whether there are any lower bounds or bottlenecks that prevent the model to find a calibration that fits the benchmark data. @LaviniaBaumstark do you have any candidates? Is this encountered in other csaes as well?

LaviniaBaumstark

please do not add soo many new config files, are all needed?

LaviniaBaumstark

Do we need to merge the realization "investment_inefficiencies" together with adding "25_WACC" module?

LaviniaBaumstark · 2026-06-22T09:10:53Z


 vm_flexAdj(tall,all_regi,all_te)                     "flexibility mark-up cost or subsidy, used to emulate price changes of technologies which see lower-than-average or higher-than-average electricity prices due to more or less flexible operation[T$/TWa]"
+
+vm_waccCost(ttot,all_regi)                               "difference between WACC costs in current and previous iteration"


Will this info be beprovided by the 25_WACC module? If so, the declaration should go to the module

If I remember correctly, when I declared it in the 25 module a check flagged an issue, because it is an interface variable, it is used in more than one module (it also appears in the budget equation). This is why I moved it there. Now, I declare it only at this point. Should I move it to the 25_module ?

yes, move it but keep the name

LaviniaBaumstark · 2026-06-22T09:16:51Z

+*** |  Contact: remind@pik-potsdam.de
+*** SOF ./modules/25_WACC/off/bounds.gms
+*fix the WACC term of the budget equation equal to zero for all times
+vm_waccCost.fx(t, regi) = 0;


please add the declariation of vm_waccCost to the 25_WACC module

LaviniaBaumstark · 2026-06-22T09:17:12Z

+*** |  REMIND License Exception, version 1.0 (see LICENSE file).
+*** |  Contact: remind@pik-potsdam.de
+*** SOF ./modules/25_WACC/off/bounds.gms
+*fix the WACC term of the budget equation equal to zero for all times


please start this line with ***

LaviniaBaumstark · 2026-06-22T09:18:14Z

  # the requested version of renv
  version <- "1.1.7"
-  attr(version, "md5") <- NULL
+  attr(version, "md5") <- "dd5d60f155dadff4c88c2fc6680504b4"


do not merge changes on the renv folder

LaviniaBaumstark · 2026-06-22T09:20:08Z


 ## [Unreleased]
-
+- **scripts** include reporttransport/remind2 (EDGE-T/REMIND) variable harmonization. reporttransport (EDGE-T side) energy service demand is rescaled to match exactly remind2 ES/FE/Emi values for variables reported by both models


looks like a merge left over. Please add your changes (e.g. add new module 25_WACC) to the CHANGELOG.md

LaviniaBaumstark · 2026-06-22T09:21:38Z

+*** |  Contact: remind@pik-potsdam.de
+*** SOF ./modules/25_WACC/standard/bounds.gms
+*fix budget equation term to zero for 2005
+***vm_waccCost.fx("2005",regi) = 0;


delete these out-commented line

LaviniaBaumstark · 2026-06-22T09:31:20Z

+    ) - p25_techwaccCostO(t, regi);
+
+***This equation calculates the capital costs for macroeconomic investments, including those coming from previous periods that are within their lifetime
+q25_invwaccCost(t, regi)$(t.val > 2005) ..


you do not need to add $(t.val > 2005) if you use t in the equation

LaviniaBaumstark · 2026-06-22T09:32:58Z

+***---------------------------------------------------------------------------
+*'  Calculation of total WACC costs
+***---------------------------------------------------------------------------
+p25_waccCostO_tewacc(t, regi, tewacc) =


Do you want to recalculate this also for fixed time periods?

LaviniaBaumstark · 2026-06-22T09:33:23Z

+
+p25_techwaccCostO(t, regi) = sum(tewacc, p25_waccCostO_tewacc(t, regi, tewacc));       
+
+*** EOF ./modules/25_WACC/standard/postsolve.gms


delete this line

LaviniaBaumstark · 2026-06-22T09:34:25Z

+*** |  REMIND License Exception, version 1.0 (see LICENSE file).
+*** |  Contact: remind@pik-potsdam.de
+*** SOF ./modules/25_WACC/standard/presolve.gms
+


if this file is empty, delete it

LaviniaBaumstark · 2026-06-22T09:35:23Z

+*** |  Contact: remind@pik-potsdam.de
+*** SOF ./modules/25_WACC/standard/sets.gms
+
+*** No new sets are directly related to this module.


I do not understand this sentence as you are adding a set

daymontas1 · 2026-06-22T10:52:21Z

Do we need to merge the realization "investment_inefficiencies" together with adding "25_WACC" module?

They are independent, so we don't have to. However, that would require creating a new branch where I remove the investment inefficiencies realization and commit only the changes in module 25. Is there an easier way to do this?

In any case, this new realization does not affect the standard realization in the macro module, and the corresponding mrremind data have already been merged. Based on these, how should I proceed?

daymontas1 · 2026-06-22T10:57:13Z

please do not add soo many new config files, are all needed?

No, I can remove them.

LaviniaBaumstark · 2026-06-22T13:09:42Z

Do we need to merge the realization "investment_inefficiencies" together with adding "25_WACC" module?

They are independent, so we don't have to. However, that would require creating a new branch where I remove the investment inefficiencies realization and commit only the changes in module 25. Is there an easier way to do this?

In any case, this new realization does not affect the standard realization in the macro module, and the corresponding mrremind data have already been merged. Based on these, how should I proceed?

In general, I would prefer having seperate PRs for different features

daymontas1 · 2026-06-22T19:57:08Z

Do we need to merge the realization "investment_inefficiencies" together with adding "25_WACC" module?

They are independent, so we don't have to. However, that would require creating a new branch where I remove the investment inefficiencies realization and commit only the changes in module 25. Is there an easier way to do this?
In any case, this new realization does not affect the standard realization in the macro module, and the corresponding mrremind data have already been merged. Based on these, how should I proceed?

In general, I would prefer having seperate PRs for different features

Ok, I will split them.

daymontas1 · 2026-06-23T09:37:15Z

Closing this in favor of PR #2378 (PR for investment inefficiencies will follow).

daymontas1 changed the title ~~Add WACC module and a new realization in the MACRO module about inves…~~ Add WACC module and a new realization in the MACRO module about investment inefficiencies Jan 29, 2026

daymontas1 requested review from LaviniaBaumstark and nicobauer January 29, 2026 11:21

daymontas1 mentioned this pull request Jan 30, 2026

Feature/phi imf efficiency pik-piam/mrremind#778

Closed

nicobauer reviewed Feb 2, 2026

View reviewed changes

Comment thread modules/01_macro/Investment_Inefficiencies/equations.gms Outdated

nicobauer reviewed Feb 2, 2026

View reviewed changes

LaviniaBaumstark requested changes Feb 3, 2026

View reviewed changes

Comment thread core/declarations.gms Outdated

Comment thread main.gms Outdated

Add WACC module and a new realization in the MACRO module about inves…

c21d420

…tment inefficiencies

daymontas1 force-pushed the imperfect-markets-wacc branch from 00e1006 to c21d420 Compare May 12, 2026 10:43

johannah-pik and others added 18 commits June 4, 2026 17:10

Update reporting.R

8395c94

include check in REMIND mif

d5cdbd2

Update reporting.R

0857bf0

Update reporting.R

3875a78

Update CHANGELOG.md

d335078

NDC delay prisma

469e7a9

debugging

f3e048c

NDC delay fix

48f8f76

ndc fix

2c818c8

ndc fix

0d0c541

delay MEA and REF to +35 years

b061b9b

special case for MEA and REF

e152abb

special case for MEA and REF

3354eb1

restore renv/activate.R

93b03bb

PRISMA config WP6

a0b8c5e

PRISMA config WP6

d3c52fa

remove new config

5fa98fc

adjust switch name

be094e5

rename invwacc to counwacc

06513a1

daymontas1 added this to the v3.0.1 milestone Jun 4, 2026

daymontas1 mentioned this pull request Jun 15, 2026

Add PHI, WACC and cost of capital files pik-piam/mrremind#822

Merged

daymontas1 and others added 12 commits June 18, 2026 10:57

Merge branch 'develop' into imperfect-markets-wacc

430923c

Add capitalcosts in the budget equation also in the standard realization

e74b09e

Merge upstream/develop, accept incoming changes

22f19c2

Apply WACC local changes on top of upstream/develop merge

43d467a

Merge branch 'imperfect-markets-wacc' of https://github.com/daymontas…

46e916d

…1/remind into imperfect-markets-wacc

Switch to the standard realization for macro module and off for 25_WACC

470c01f

Merge branch 'develop' into imperfect-markets-wacc

c0ea50b

Remove preloop.gms from WACC standard module

f84a6be

Merge branch 'imperfect-markets-wacc' of https://github.com/daymontas…

ada0916

…1/remind into imperfect-markets-wacc

Found the issue with investment inefficiencies in past investments

dd475e7

Remove stale tax-related variable declarations

d377281

Mark pm_ts as not used in WACC off realization

ba8e578

remindmodel deleted a comment from LaviniaBaumstark Jun 18, 2026

daymontas1 requested a review from LaviniaBaumstark June 18, 2026 10:37

LaviniaBaumstark reviewed Jun 22, 2026

View reviewed changes

daymontas1 added 2 commits June 22, 2026 22:30

WIP: latest WACC module changes

44e5074

Commit latest changes before split the PR

2e8bf35

daymontas1 closed this Jun 23, 2026


		vm_flexAdj(tall,all_regi,all_te) "flexibility mark-up cost or subsidy, used to emulate price changes of technologies which see lower-than-average or higher-than-average electricity prices due to more or less flexible operation[T$/TWa]"

		vm_waccCost(ttot,all_regi) "difference between WACC costs in current and previous iteration"


		## [Unreleased]

		- scripts include reporttransport/remind2 (EDGE-T/REMIND) variable harmonization. reporttransport (EDGE-T side) energy service demand is rescaled to match exactly remind2 ES/FE/Emi values for variables reported by both models


		p25_techwaccCostO(t, regi) = sum(tewacc, p25_waccCostO_tewacc(t, regi, tewacc));

		*** EOF ./modules/25_WACC/standard/postsolve.gms

Conversation

daymontas1 commented Jan 29, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Purpose of this PR

Description

Parts concerned

Impact

Checklist

Further information (optional)

Uh oh!

daymontas1 commented Jan 30, 2026

Uh oh!

Uh oh!

nicobauer left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

daymontas1 commented Mar 26, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

LaviniaBaumstark commented Mar 30, 2026

Uh oh!

daymontas1 commented Jun 15, 2026

Uh oh!

nicobauer commented Jun 18, 2026

Uh oh!

LaviniaBaumstark left a comment

Choose a reason for hiding this comment

Uh oh!

LaviniaBaumstark left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

daymontas1 Jun 22, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

daymontas1 commented Jun 22, 2026

Uh oh!

daymontas1 commented Jun 22, 2026

Uh oh!

LaviniaBaumstark commented Jun 22, 2026

Uh oh!

daymontas1 commented Jun 22, 2026

daymontas1 commented Jan 29, 2026 •

edited

Loading

daymontas1 commented Mar 26, 2026 •

edited

Loading

daymontas1 Jun 22, 2026 •

edited

Loading