# Where ₹100 goes in rural and urban India Original Meplar chart and source transcription, prepared 17 October 2026. Insight: https://meplar.com/insights/india-household-spending/ Dataset: https://meplar.com/datasets/india-household-spending/ ## Main comparison Of every ₹100 of average monthly consumption expenditure per person, food accounts for ₹47.04 in rural India and ₹39.68 in urban India. Non-food is ₹52.96 and ₹60.32 respectively, so each displayed sector bar totals ₹100. The rural food share is 7.36 percentage points higher. Yet average food consumption expenditure is higher in urban India: ₹2,776 per person per month, versus ₹1,939 in rural India. The corresponding total MPCE averages are ₹6,996 and ₹4,122. A smaller percentage can therefore accompany a larger rupee amount. These are published rural and urban averages, rather than the same households followed over time. The measure includes valued home-grown consumption and gifts, so it is broader than money paid out. Section I leaves out the additional valuation of specified free welfare goods; it does not remove every kind of imputation. ## Scope and definitions All-India rural and urban averages · August 2023–July 2024 · HCES Section I. Shares of average monthly consumption per person, not income. Home-grown consumption and gifts are included; the additional valuation of specified free welfare goods is excluded. This is average Monthly Per Capita Consumption Expenditure (MPCE), using the source's published nominal consumption estimates. It is not household income, median expenditure, a household total, cash-only spending or a causal effect of living in a rural/urban area. ₹100 is a common display denominator for composition, not either sector's actual monthly total. The official Section I basis excludes additional values of specified goods received free through social welfare programmes (printed page 3 paragraph b). It retains ordinary valuation of consumption from home-grown/home-produced stock, gifts, loans, free collection and goods exchanged (paragraph a). Free health/education service values are not imputed in the source. Do not shorten the basis to an unqualified no-imputation claim. ## Two food groups, already within the food total | Published group | Rural share of TOTAL MPCE | Urban share of TOTAL MPCE | |---|---:|---:| | beverages, refreshments, processed food (includes purchased cooked meals) | 9.84% | 11.09% | | cereals & cereal substitutes | 4.99% | 3.76% | These are shares of total MPCE, not percentages within food and not extra chart slices. The broad first group must not be relabelled packaged, ultra-processed or junk food. The separate pulses group includes gram. ## Method and limits Transcribe all 24 Statement 4 rows from the official HCES 2023–24 Fact Sheet and check every label and all 96 numeric cells against both PDF text and the rendered table. Preserve the published whole-rupee MPCE amounts and two-decimal share strings. Select only food total and non-food total for each sector; express each published percentage as rupees per ₹100 of total average monthly per-capita consumption expenditure. The two segments sum to ₹100 in each sector. Do not add subtotals to their component rows or recompute shares from the displayed rupee amounts. - MPCE means average Monthly Per Capita Consumption Expenditure. It is not income, median expenditure, a household total or the budget of a typical family. - The survey period is August 2023 through July 2024, not an April–March fiscal year. Values are nominal Indian rupees; no inflation adjustment is added. - Section I excludes the additional imputed values of specified goods received free through social welfare programmes. It retains valuation of home-grown/home-produced consumption, gifts, loans, free collection and goods received in exchange. Do not describe this as no imputation or cash-only spending. - A smaller urban food share accompanies a higher urban rupee amount: ₹2,776 per person per month versus ₹1,939 in rural India. The total MPCE denominators are ₹6,996 and ₹4,122 respectively. - The supporting food-group figures are shares of total MPCE and are already included in the food segment. The broad beverages, refreshments, processed food group includes purchased cooked meals; it is not just packaged or ultra-processed food. Cereals include cereal substitutes; the pulses group includes gram. - Different source items use different recall periods under the Modified Mixed Reference Period approach; the published results are expressed as monthly MPCE. The factsheet does not supply a full household-level estimator here, and Meplar does not invent one. - All-India survey coverage excludes a few inaccessible villages in the Andaman and Nicobar Islands. These are published survey estimates, without uncertainty intervals shown in the graphic. - This single-period rural/urban comparison does not establish a time trend or the causes of consumption differences. No affordability, health, income or urbanisation effect is estimated. ## Download files - chart_data.csv: Four plotted sector/category rows, with published shares and their MPCE denominators. - statement4.csv: All 24 Statement 4 rows, including explicitly marked subtotals. - chart.png: Reviewed original chart, 1440 × 2658 pixels. - chart.svg: Self-contained outlined vector chart; no external font dependency. - README.md: Interpretation, scope, fields, source provenance and offline instructions. - SOURCE_TERMS.txt: Verified MoSPI source-reuse permission and conditions. - SOURCE_PROVENANCE.json: Official source locations, hashes, selection and reuse evidence. - table-extraction-notes.json: Exact table extraction and rounding checks. - source-reuse-record.json: Compact verified Category A reuse record; no unrelated website bundle. - reproduction.zip: Standalone offline build with frozen CSV, original factsheet PDF, code and source notes. - SHA256SUMS.txt: SHA-256 hashes of all other public files in this directory. `chart_data.csv` has four rows and columns sector, category, share_pct, rupees_per_100_average_mpce, category_mpce_rs and total_mpce_rs. The notional rupees-per-100 column equals the source percentage numerically; it is not a new survey estimate. `statement4.csv` preserves all 24 rows and has source_row, item_group, category_kind, broad_category, rural_mpce_rs, urban_mpce_rs, rural_share_pct, urban_share_pct and footnote_marker. The 21 group rows, two subtotals and all-items total overlap; do not sum every row. Published percent strings retain two decimals and rupee fields retain their whole-rupee display precision. Some shares differ from ratios of the displayed rupees, so the published share column is authoritative for this graphic. ## Reproduce offline Extract `reproduction.zip`, open its `reproduction/` directory and run `python3 build_chart.py` in a Python environment that already has Matplotlib 3.9.4. The script makes no network request and performs no installation. It checks the frozen 24-row CSV hash and the exact four plotted share strings before rendering. Outputs go to overlay/downloads/india-household-spending/v1/, exports/ and checks/. The original unchanged official factsheet PDF, frozen CSV, extraction notes, source provenance, compact reuse record and source-extraction script are included. The chart build reads the CSV; Poppler's `pdftotext` is needed only if rerunning sources/extract_statement4.py to check the table against the original PDF. Re-running extraction revalidates all 24 labels and 96 source numeric cells. The reviewed chart uses installed Arial with Matplotlib's bundled DejaVu Sans per-glyph fallback for ₹, and saves outlined SVG type. No font files are redistributed. Numeric outputs are independent of font substitution; different installed fonts/Matplotlib versions can change pixel/outline hashes. Set MEPLAR_FONT_FAMILY only when intentionally choosing another installed font. Reviewed PNG/SVG/CSV outputs and native 390px raster are included for comparison. Automated checks cover published-share sums, subtotal selection, the ₹ glyph, minimum label sizes, and clipped/overlapping text. The actual 390×720 PNG was independently inspected. Essential chart labels are at least 14.25px, numeric values 20px and source-only credits 11px at that display size. No browser is used by the offline build. ## Source and reuse SOURCE DATA AND OFFICIAL FACTSHEET Ministry of Statistics & Programme Implementation / National Statistical Office (MoSPI / NSO), Government of India. HCES 2023–24 Fact Sheet, Statement 4, printed page 8 (PDF page 9); Section I definitions on printed pages 3–4 (PDF pages 4–5). Official factsheet: https://www.mospi.gov.in/sites/default/files/publication_reports/HCES%20FactSheet%202023-24.pdf VERIFIED REUSE BASIS Official MoSPI copyright policy: https://www.mospi.gov.in/CopyrightPolicy The policy permits accurate reproduction and allows NSO/MoSPI Category A datasets to be accessed, reproduced, redistributed, adapted and reused for commercial or noncommercial purposes, subject to appropriate attribution and other applicable legal conditions. The source must be prominently acknowledged wherever the material is published or issued to others. This permission does not cover separately identified third-party copyright material; obtain authorization from its holder where applicable. GSDD 2026: https://www.mospi.gov.in/uploads/acts_and_policies/acts_and_policies1778231030122_a1a3ee77-e36d-4895-8da5-b294751cbb1a_272304_(1).pdf English printed page 13, clause A.3.1 expressly includes NSS reports and factsheets in open-access aggregated/analyzed Category A publications. Clause 5.7.1 provides Category A access free without registration. This published HCES factsheet table fits that explicit class; it is not Category B unit-level microdata or restricted Category C data. The applicable confidentiality/privacy and other legal conditions remain in force. Official policy response SHA-256: ecf20f078cb3adf41f1d303cabfa91ce66d90d6ba8d11f013f09814583448852 Official GSDD 2026 PDF SHA-256: 6503121c5f7a08d0bf6a8b2ecf618a30c40b3b5a5d3dc53b3fc339fa434d32c8 Frozen HCES factsheet SHA-256: 31d4ce93ce9396b4042d226408b6d12e4be2006e0b716a0d63995d913b13f76e The exact compact public-policy response was archived locally and independently reviewed; the GSDD PDF remains private evidence. No unrelated public-website application bundle is included in this package. HANDLING AND CHANGES Retain prominent MoSPI/NSO attribution, source links, table labels/footnotes, survey period, nominal monthly-per-person units and the distinction between ordinary in-kind valuation and additional specified free-welfare-item imputation. Published percentages and whole-rupee figures are retained without inventing more precision. Meplar selected the food and non-food subtotals for an original ₹100 composition graphic and retained all 24 source rows separately. The original factsheet PDF is unchanged in the offline reproduction package. No CC0, GODL, Creative Commons or public-domain designation is assigned. No Meplar terms relicense the official source document. Original Meplar chart artwork/code are separate works; no official endorsement is implied. No separately identified third-party artwork is reused in the original chart. ## Tools and affiliation Original Meplar chart/design with AI assistance from ChatGPT/Codex. Tools: Python standard library, Decimal arithmetic, Matplotlib 3.9.4 and local Poppler for PDF extraction. I run Meplar. No paid assets or services were used. Additional external cost: ₹0. No source endorsement is implied.