Key Takeaways
- Land is the widest spread. Sampled GIDC estates run from $8.99 per sq m (Sanand, RIDC) to $115.22 (Naroda and Vatva) in FY2026–27, about 12.8 times. SIPCOT listings run from $16,533 to $211,835 per acre, also about 12.8 times (NextMSC calculations).
- Construction benchmarks are tight. JLL’s 2025 guide puts light and medium industrial buildings at $68.20–$83.70 per sq ft across six cities and heavy industrial at $90.93–$112.63, a city spread of roughly 23–24%. Chennai is lowest and Mumbai highest in both.
- Power ranges overlap heavily. Indicative HT landed midpoints run from $0.081 per kWh (Odisha) to $0.100 (Maharashtra). At 100,000 kWh a month, that gap is about $1,900 a month, or $22,900 a year (NextMSC calculation). These are secondary screening figures, not tariff orders.
- Labour and incentives cannot yet rank states. Six of the 12 states in scope have no comparable Labour Bureau wage figure, and incentive packages are conditional, capped or alternative, so they are not additive.
Market Context: Why Cost by State Needs a Decomposition
A single “cheapest state” answer does not survive contact with the data. The cost of a factory has a one-time part (land, building, connections) and a recurring part (power, labour, taxes), and each is quoted on a different basis: $ per sq m or per acre for land, $ per sq ft for buildings, $ per kWh or per kVAh for power and $ per day for labour. All dollar values in this analysis are Indian rupee figures converted at INR 96.7733 per USD (FBIL reference rate, 8 October 2026); official tariffs, wages and subsidies remain denominated in rupees. Incentives then reduce some of these costs only for eligible projects.
This Deep Dive therefore compares one cost at a time and states what each figure does and does not measure. It builds on NextMSC’s earlier work on state semiconductor and electronics incentives.
Labour: A Proxy, Not a Payroll Floor
The only comparable official series in this dataset is the Labour Bureau’s January 2026 rural wage survey for non-agriculture construction workers. It covers six of the 12 states in scope (Andhra Pradesh, Assam, Bihar, Gujarat, Haryana and Karnataka). Male average daily wages run from $3.93 in Bihar to $6.59 in Karnataka, about 68% higher (NextMSC calculation). Himachal Pradesh, outside the 12, reports $6.67.
Construction-Worker Average Daily Wage, Male, January 2026 ($/day)
Labour Bureau rural wage observations for non-agriculture construction workers; a construction proxy, not a factory payroll floor. The highest value is highlighted.
These are construction-sector wages. They do not equal statutory minimum wages or factory operator pay, and Maharashtra, Tamil Nadu, Telangana, Uttar Pradesh, Odisha and Rajasthan have no comparable figure here. JLL also projects that the November 2025 Labour Codes could raise construction labour costs by 5–12% in 2026, a projection rather than a measured change.
Power: Overlapping Ranges, Different Units
Industry-published HT landed tariff ranges give a screening view. Midpoints run from $0.081 per kWh in Odisha to $0.100 in Maharashtra, with Gujarat ($0.087) and Andhra Pradesh ($0.088) closer to the lower end. At a sample load of 100,000 kWh a month, monthly energy spend at the midpoint ranges from $8,112 in Odisha to $10,023 in Maharashtra (NextMSC calculation).
Indicative HT Landed Power Tariff by State ($/kWh, low–high)
Secondary industry screening ranges, not tariff orders. Solid bar = low end; lighter bar = high end. Lowest midpoint is highlighted.
The only official tariff example in the dataset is Haryana’s HERC schedule for FY2026–27, which sets industrial HT energy at $0.0708 per kVAh (load factor up to 50%) and $0.0682 per kVAh (above 50%), with a 22.5% solar-hours rebate and a 7.5% continuous-supply surcharge where applicable. A kVAh rate is not a landed $ per kWh rate, so the two should not be compared directly. Bills also depend on demand charges, time-of-day rules, duties and open-access options.
Construction: The Narrowest Spread
Building cost varies least across cities. The table sets JLL’s 2025 benchmarks side by side. They are building costs in $ per sq ft, exclude land, process equipment, external works, utilities, financing and contingency, and a warehouse is not a process factory.
Source: JLL Construction Cost Guide India 2025 and NextMSC analysis. Ranges are low–high benchmarks in $ per sq ft (INR converted at 96.7733 per USD).
JLL’s 2026 guide adds three outlook items: possible GST 2.0 savings of 2–5% in construction cost, and steel, aluminium and copper cost rises of 2–4%, alongside the labour-code effect above. All three are projections and should be treated as sensitivity ranges.
Land: Where the Real Dispersion Sits
Official estate listings show the widest variation, and it occurs inside states as much as between them. Gujarat’s sampled GIDC rates span $8.99 per sq m at Sanand (RIDC) to $115.22 at Naroda and Vatva, with Bhuj at $9.20 and Gandhidham at $106.64. Tamil Nadu’s SIPCOT samples span $4.09 per sq m (Cuddalore Phase III) to $52.35 (Aerospace Park, Vallam Vadagal), equivalent to $16,533–$211,835 per acre. Odisha’s three IDCO samples, from a circular dated 8 November 2024 that must be reconfirmed, fall between $8.27 and $16.97 per sq m.
Sampled Industrial Land Rates by State ($ per sq m, lowest–highest sample)
Official estate listings converted to $/sq m; samples are not state averages. Solid bar = lowest sample; lighter bar = highest sample.
Source: GIDC, SIPCOT, IDCO and NextMSC analysis. All rates are shown as $ per sq m; SIPCOT per-acre listings are converted arithmetically. Maharashtra (MIDC) and Andhra Pradesh (APIIC) publish plot-level lookups, so no state rate is shown.
These are samples, not state averages, and they exclude development charges and lease or sale terms. The practical reading is that estate choice matters more to the land budget than state choice, a point that also frames the wider industrial real estate market.
Incentives: Conditional Upside, Not a Discount
State packages differ in form as much as in size: capital subsidies, SGST reimbursement, power reimbursement, stamp-duty relief and turnover-linked support. Percentages are rarely comparable, because some are ceilings, some are caps on annual reimbursement and some are alternatives. The table sets out what each policy states.
Source: Official state policy documents and NextMSC analysis. Confirm current operational guidelines, eligibility, approval dates and scheme caps before using any benefit in a business case.
Three patterns stand out. Andhra Pradesh, Odisha and Maharashtra express support against capital (15% of eligible fixed capital investment, 20–30% of plant and machinery, and ceilings of 30–100% of fixed capital investment respectively). Rajasthan and Uttar Pradesh lean on tax and duty relief. Odisha and Tamil Nadu add recurring power relief, about $0.0207 per kWh for 7 or 10 years in Odisha and a five-year electricity tax exemption in Tamil Nadu.
Original Comparison: How Far Can Each Cost Rank States?
The table below is NextMSC’s synthesis of the verified ranges. It shows how wide each cost spread is and how safely it can be used to rank states.
Source: Labour Bureau, JLL, GIDC, SIPCOT, Pro Solar EPC and NextMSC analysis. Spreads are highest value divided by lowest value within each sample (NextMSC calculation).
Illustrative Scenario: A 50,000 sq ft Plant
For a plant of 50,000 sq ft on five acres using 100,000 kWh a month, the workbook’s screening model gives three formula results. The building estimate is $3.46 million at the $69.23 per sq ft low-end Bengaluru benchmark, or $3.41–4.19 million across the full JLL light and medium range (NextMSC calculation). Annual energy cost is $91,761 to $126,481, using $0.0765 to $0.1054 per kWh. A construction-phase labour proxy of 20 workers for 180 days at $6.03 a day adds about $21,713.
These figures give a gross subtotal proxy of about $3.48 million. They exclude land, machinery, utilities, taxes, financing, contingency and incentives, so they are a floor for comparison and not a project budget.
Implications for Industry Stakeholders
Site-selection and capex teams
Because land varies more inside a state than between states, shortlist estates before states, and request the full charge schedule from the relevant authority. Construction cost should not decide the location.
Finance and operations teams
Model recurring power from the applicable tariff order and load profile, not from a state midpoint, and carry incentives as conditional upside with their caps and alternatives stated.
Policy and investment-promotion teams
The data show comparable numbers exist for land and construction but not for factory labour or power. Publishing consistent, dated schedules would make state comparisons more reliable.
Methodology, Definitions and Limitations
Data cut-off: 9 October 2026. All dollar values are INR converted at 96.7733 per USD (FBIL reference rate, 8 October 2026), a conversion convention and not a forecast of exchange rates. Wages are January 2026 observations, JLL construction benchmarks are from the 2025 guide, land rates are FY2026–27 listings (Tamil Nadu as checked on 9 October 2026, Odisha from a circular dated 8 November 2024) and incentive policies run from 2021 to 2029. Because the series have different dates, the period is shown beside each metric. Planned refresh: quarterly, preserving field definitions.
Source hierarchy
Official sources come first: Labour Bureau rural wage observations, the CEA tariff and duty report (March 2026) and the HERC FY2026–27 schedule, GIDC, SIPCOT and IDCO land listings, and official state policy documents. Industry sources follow: JLL’s 2025 construction cost guide and 2026 outlook, and Pro Solar EPC’s indicative HT tariff ranges, which are secondary screening estimates with an unstated effective date. Where an official figure is on a different basis, such as Haryana’s $ per kVAh, it is shown separately and not substituted for a landed $ per kWh range.
Data types
Reported: presented directly by the source under its own definition, such as wages, JLL ranges, land listings and policy terms. Calculated: derived arithmetically from reported values, such as spreads, midpoints, normalised $ per sq m land rates and the scenario outputs. Proxy: a measure standing in for the quantity of interest, such as construction-worker wages for factory labour cost. Secondary estimate: an industry-published figure without an official order, such as the HT tariff ranges. Projection: a forward-looking source view, such as JLL’s labour-code, GST and metals outlook.
Limitations
Labour data cover six of the 12 states in scope and describe construction work, not factory payroll. Power ranges are landed screening figures; bills depend on contract demand, voltage, load factor, time-of-day rules, duties and open access. Land figures are estate samples, not state averages, and exclude development charges and lease or sale terms. Construction benchmarks cover the building only. Incentives are conditional ceilings or alternatives and are not additive.
The workbook’s quick-findings note gives $6.59 (Karnataka) as the top male construction wage, but its own table lists Himachal Pradesh at $6.67, so this article states the range for the six manufacturing states with data and notes Himachal Pradesh separately. Corrections: contact NextMSC research.
Project Dataset Explorer
Filter the labour, power, construction, land and incentive datasets compiled for this report.
Outlook: What to Monitor
NextMSC analysis, not a forecast. Watch: the effect of the Labour Codes on construction wages; GST 2.0 and metal prices in building costs; current SERC tariff orders for FY2026–27; Odisha’s amended industrial policy text; the end of Gujarat’s Aatmanirbhar assistance window on 4 October 2027; and live plot rates from MIDC and APIIC, which would close the biggest gaps in this comparison.
Conclusion: What the Cost Data Show
Factory cost by state is best read as five separate questions rather than one league table. The evidence supports clear statements about where costs differ and by how much, and equally clear limits on what it can rank.
What the data establish
Construction cost is the most uniform input, with a city spread of about 23–24% in JLL’s 2025 benchmarks. Land is the most dispersed, at about 12.8 times within both Gujarat’s sampled GIDC estates and Tamil Nadu’s SIPCOT listings. Indicative power midpoints differ by about 24% across ten states, and construction-worker wages differ by about 68% between Bihar and Karnataka.
What the data do not establish
They do not identify a cheapest state. Factory labour cost, official power bills and plot-level land prices are not available on a common basis, and incentives are conditional ceilings, caps or alternatives that reduce eligible costs only after approval.
How to use the comparison
Shortlist estates before states, price recurring power from the applicable tariff order and load profile, and carry incentives as conditional upside with their caps stated. Refresh the inputs as the Labour Codes, GST changes, metal prices and amended state policies take effect.
Data cut-off: October 9, 2026 · Corrections and source challenges: Contact NMSC Research.
About the Author
Debashree Dey
— Debashree Dey is Assistant Manager at Next Move Strategy Consulting, where she supports cross-vertical market content and communications across diverse industries for 6 years. Her professional background includes senior content writing, communications, and published manuscript authorship, with experience developing audience-focused business narratives and maintaining clear, consistent messaging. Her role supports research-led content development and editorial quality across NextMSC publications.
About the Reviewer
Supradip Baul
— Supradip Baul is an accomplished business consultant and strategist with over a decade of rich experience in market intelligence, strategy, technology, and business transformation. His work has included rigorous qualitative and quantitative analysis across multiple industries, helping clients shape investment decisions and long-term roadmaps. Earlier in his career, he was associated with Gartner, where he contributed to industry-leading reports and market share analyses. He has worked with leading global companies and holds an MBA with a dual specialization in Marketing and Finance.