Revised 23 October 2026 · v1.0.0 · what changed

Code: github.com/herrrickshaw/india-forex-analysis

Watch the walkthrough

Watch on YouTube: youtu.be/2yPEcJLO08s

Why it is useful

• Who benefits: students and policy readers who want one worked narrative on reserves, the trade balance and debt service, with the working shown in plain markdown tables.
• What it replaces: stitching together RBI reserve tables, Budget annexes and TradeStat extracts by hand for a first-pass picture.
• Evidence from our read: the reserve table adds up (552.28 + 115.40 + 18.62 + 4.81 = 691.11) and the tax-share percentages recompute correctly.
• Evidence of a problem: the stated 2023 exports of US$875.4 bn are exactly twice the 437.7 bn total row in the sibling india-trade-export-analysis data, so the headline deficit looks unreliable.
• Treat it as a worked example to audit, not a source to cite.

What it is for

The repository is nine numbered markdown documents plus an index. They cover export growth 2020-2023, import and export analysis, RBI reserves versus the trade deficit, a reconciliation of Budget 2026-27 tax revenue, debt and borrowing trends, a five-year post-pandemic comparison, year-on-year deterioration, multi-source integration and an executive summary. There is no code to run.

It rates the situation an 'amber alert' and gives scenario probabilities. Those ratings and probabilities are the author's judgement, not outputs of a documented model. This is economic commentary for learning, not investment, trading or policy advice. The repository is archived (read-only).

Who it is for

• Students of macroeconomics practising how to audit a report
• Journalists who want a checklist of numbers to verify before quoting
• Analysts looking for a structure to rebuild with sourced data

What you need

Get the code and an AI assistant

You do not have to be a programmer. Get the repository, give it to an AI assistant, and ask in plain language. Any of these routes works:

git clone https://github.com/herrrickshaw/india-forex-analysis.git
cd india-forex-analysis
claude      # or: gemini

Install

1. Get the documents

   git clone https://github.com/herrrickshaw/india-forex-analysis.git
   cd india-forex-analysis

2. List what is there

   ls *.md

Quick start: commands and what you should see

Start with the index and README, then the reserves document

less README.md
less 3_RBI_RESERVES_vs_TRADE_DEFICIT_ANALYSIS.md

You see the executive summary, the amber-alert framing and the reserve composition table.

Re-compute the stated figures

python3 -I check.py   # a short script that re-adds the tables you copy from the documents

Five lines of arithmetic checks, shown in the sample outputs below. Our check script is not part of the repository.

Search for a number across all documents

grep -n "875.4" *.md

The same figure appears in several documents, which makes a single error spread.

Sample outputs

Real output from a run on the date shown in the sources line; trimmed to fit.

Arithmetic checks on figures the documents state

python3 -I check.py
reserve components sum: 691.11 (stated 691.11)
Sep-2025 to Mar-2026 change: -8.98 bn (-1.28%)
trade deficit monthly drain: 2023 47.2, 2026 64.1 bn
2026 deficit / 2019-20 deficit: 7.33x (stated 7.3x)
tax shares: Corp 27.9%, Income 33.3%, GST 23.1%, Excise 8.8%, Customs 6.2%, Other 0.6%

These pass: reserve components, the six-month change, the monthly-drain division and the tax shares are internally consistent.

Where the export figure goes wrong

grep -n '875.4' 2_IMPORT_EXPORT_ANALYSIS_2025_26.md; then read the 'India's Total Export' row of the sibling repo CSV
2_IMPORT_EXPORT_ANALYSIS_2025_26.md:13:- **Exports**: $875.4B
2_IMPORT_EXPORT_ANALYSIS_2025_26.md:79:Export Growth (2020-2023):       +44% ($583.6B → $875.4B) ✓ Strong
total row 2020: 291808.48  2023: 437704.58 (USD mn)   <- india-trade-export-analysis CSV
2 x total row: 583616.96  875409.16

US$875.4 bn is 2 x 437.7 bn. The sibling data has one 'India's Total Export' row plus the commodity rows, so adding both appears to have double-counted. The imports figure may have the same problem, which we did not check, so the stated deficit cannot be trusted in either direction until it is re-derived.

Reserve composition as stated (March 2026)

grep -n -A7 'Asset Class' 3_RBI_RESERVES_vs_TRADE_DEFICIT_ANALYSIS.md
| Asset Class | Value ($B) | % of Total |
| Foreign Currency Assets | 552.28 | 79.9% |
| Gold (880.52 MT) | 115.40 | 16.7% |
| SDRs | 18.62 | 2.7% |
| IMF Reserve Position | 4.81 | 0.7% |
| TOTAL | 691.11 | 100% |

Quoted from the repository; we did not check it against the RBI release.

Ways to use it

Sample prompts for your AI assistant

Copy, paste and edit the parts in [brackets].

• Read README.md and 2_IMPORT_EXPORT_ANALYSIS_2025_26.md in this repo and list every number that cannot be derived from another number in the repo.
• Recompute the reserve composition percentages in 3_RBI_RESERVES_vs_TRADE_DEFICIT_ANALYSIS.md and tell me if any do not add up.
• The repo says exports were US$875.4 bn in 2023. Compare with [the official figure from the Commerce Ministry] and explain the gap.
• Rewrite the executive summary so every claim states its source and whether it is actual, estimated or projected.
• List the assumptions behind the scenario probabilities (25/60/15) and say what evidence would be needed to justify them.

Limits and honest caveats

Sources: Repository herrrickshaw/india-forex-analysis (archived, read on 2026-10-23); cross-check against herrrickshaw/india-trade-export-analysis EXTENDED_ANALYSIS_2020_2026.csv; arithmetic run with Python 3 standard library on 2026-10-23. Not run: nothing else (no code in repo).

About this article: Researched, written and edited by Umashankar Triplicane Dwarakanathan, with AI research assistance; every figure is meant to trace to the primary source cited. See the Editorial Policy for how sourcing, AI use and corrections work.