Every calendar year of DPIIT's Industrial Entrepreneurs Memoranda since 2017: how much investment intent got FILED (Part A) vs how much actually got BUILT (Part B) — the closest thing India's manufacturing-investment data has to a government report card.
IEM Conversion — A Historical Performance Review (2017–2025)
The headline number: is conversion improving?
Yes, and sharply — but read the pipeline-effect caveat before treating any single year as a report card. Conversion (Part B implemented ÷ same-year Part A filed) is a same-period flow ratio, not a cohort match: this year's Part B often implements filings from 2-3 years earlier, so a high ratio can mean either "government executing faster" or "an old backlog finally landing." Both readings matter for "how is the government performing," just differently.
| Calendar Year | Part A Filed (Rs cr) | Part A (IEMs) | Part B Implemented (Rs cr) | Part B (IEMs) | Conversion % | Note |
|---|---|---|---|---|---|---|
| 2017 | — | — | 71,396 | 571 | — | |
| 2018 | 458,652 | 2,173 | 255,783 | 1,005 | 55.8 | |
| 2019 | 678,852 | 2,378 | — | 1,229 | — | 2019 Part B EXCLUDED — source defect: AR2022-23 prints Gujarat 2019 at Rs 15,18,861cr (~6x all-India); AR2021-22's 2019 column repeats the IEM count in the investment cell |
| 2020 | 414,620 | 1,432 | 233,969 | 848 | 56.4 | |
| 2021 | 771,428 | 1,489 | 347,978 | 812 | 45.1 | |
| 2022 | 424,101 | 1,185 | 250,064 | 916 | 59.0 | |
| 2023 | 510,852 | 1,267 | 276,925 | 968 | 54.2 | |
| 2024 | 602,082 | 986 | 493,765 | 1,165 | 82.0 | |
| Up to Dec 2025 | 530,416 | 702 | 583,096 | 998 | 109.9 | part-year, revises heavily |
Reading the arc, year by year
- 2018-2020 (pre-pandemic baseline): conversion holding in the mid-50s (55.8%, then 2019 excluded for a source defect, 56.4% in 2020) — a fairly steady execution rate even through COVID's first year.
- 2021 (45.1%, the low point): Part A filing surged to Rs 7.71L cr (post-COVID investment optimism) but Part B implementation lagged — construction/commissioning delays from the pandemic disruption showing up with a lag.
- 2022-2023 (rebound then slip, 59.0% → 54.2%): conversion rebounds off the 2021 low but slips back the following year, still not breaking out — a normalising-but-not-yet-accelerating execution environment.
- 2024 (82.0%, the inflection): Part B implementation jumps to Rs 4.94L cr against a Part A base of just Rs 6.02L cr — both because 2024's own filings converted faster AND because the 2021-22 filing surge is landing now. This is the year IEM implementation (Rs 4,93,765cr, CY2024 as reconciled in the Dec-2025 Annual Report vintage) overtook FY2024-25 FDI equity inflow (Rs 4,21,929cr) for the first time — domestic industrial capex outpacing foreign investment as a growth engine (calendar-year IEM vs. fiscal-year FDI: the periods overlap but aren't identical, since DPIIT reports each series on its own native calendar).
- "Up to Dec 2025" (109.9%, provisional): conversion above 100% is mechanically a pipeline effect (this year's Part B still draws on earlier Part A filings while 2025's own Part A count is undercounted mid-year) — not read as a literal "more built than promised" signal. DPIIT's own partial-year figures have historically revised upward between consecutive Annual Reports (magnitude not independently verified for this piece), so treat this row as a floor, not a final figure.
What this does and doesn't tell you about government performance
Supports a genuine improvement story: the 2024 inflection isn't just a pipeline artefact — the raw implemented value (Rs 4.94L cr) is the highest single year on record, and it happened while the Part A base (Rs 6.02L cr) sat well below 2021's outlier peak (Rs 7.71L cr), meaning most of 2024's Part A filings weren't just filed but built. That is a genuinely stronger execution signal than 2021's big-filing/weak-conversion year.
Caveats that keep this from being a clean report card:
- Conversion is a same-period FLOW ratio, not a cohort match — Part B in any year draws on Part A filed in earlier years, so "this year's conversion rate" is not "the government converted this year's promises."
- Capital source is not recorded — a state-incentive-driven private capex cycle (steel, chemicals, cement) can move this number independent of any specific central policy action.
- The eligibility threshold changed (Rs 50cr → Rs 125cr from Apr 2025) mid-series — later years structurally exclude smaller projects that would have counted in 2017-2024, so year-over-year IEM counts (not Rs cr values) are not fully comparable across the threshold change.
- 2019 is excluded entirely for a documented DPIIT source defect (Gujarat printed at ~6x plausible value in one Annual Report vintage) — a reminder that even the "official" series needs cross-version reconciliation, not blind trust in the latest Annual Report.
Where the 2024 acceleration is landing
State-level Part B data (see the companion IEM State & Sector Analysis) shows the 2024-2025 jump concentrated in a few states: Gujarat alone implemented Rs 4.55L cr across just 2024-2025, and the top-3 states (Gujarat, Odisha, Maharashtra) account for 63.3% of all 2021-2025 implementation. The national conversion-rate story is thus not evenly distributed — it is disproportionately a story about a handful of large industrial states executing faster, not a uniform national acceleration.
Source: DPIIT Annual Reports 2018-19 through 2025-26 (differenced/reconciled across vintages), cross-checked against FDI equity inflow data. Built 22 Jul 2026, published 2 Aug 2026. Companion: Where India's Factories Actually Get Built — IEM State & Sector Analysis. Research synthesis — not investment or policy advice.
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.
