Are Indians Borrowing More and Investing Differently Than 20 Years Ago?
Yes—but not in one smooth, annual series. New household liabilities rose from roughly ₹27 per ₹100 of new financial assets in 2004–05 to about ₹55 in 2023–24. At the same time, deposits remain the largest destination for new financial saving, while pension, insurance, mutual funds and SIPs make the financial mix broader. Consumption has shifted away from food—especially cereals—but rural and urban baskets still differ.
Household balance sheets are more leveraged—and more financially varied
The clearest comparable flow points show liabilities absorbing a much larger share of newly acquired financial assets after 2021–22. But “Indians are borrowing more” is not the same as “more households are indebted”: RBI bank-credit totals, national-account flows and AIDIS survey incidence measure different things.
For household savings, the strongest annual macro measure here is net financial assets as a share of GDP—not a survey of whether each family saved in a given year.
New liabilities equalled 6.4% of GDP in 2023–24, versus gross financial assets of 11.6%.
Deposits still dominate flows, but pension, insurance, small savings and market products each matter.
Food’s share fell over two decades, while non-food categories now take the majority of MPCE.
India Household Money Shift Explorer
Switch lenses without mixing their units or source years.
Flow measure derived from RBI household financial accounts. Missing years stay N/A where this page could not establish a comparable GDP-denominator pair.
The liability side changed fastest after the pandemic-era saving spike
| Year | Gross financial assets | Financial liabilities | Net financial assets | Liabilities per ₹100 assets |
|---|---|---|---|---|
| 2004–05 | 13.79% of GDP | 3.70% of GDP | 10.09% of GDP | ₹26.8 |
| 2009–10 | N/A | N/A | N/A | N/A — not silently interpolated |
| 2014–15 | N/A | N/A | N/A | N/A — source denominator differs |
| 2019–20 | 10.6% of GDP | 2.9% of GDP | 7.7% of GDP | ₹27.4 |
| 2021–22 | 11.1% of GDP | 3.8% of GDP | 7.3% of GDP | ₹34.4 |
| 2022–23 | 10.9% of GDP | 5.9% of GDP | 5.0% of GDP | ₹54.4 |
| 2023–24 | 11.6% of GDP | 6.4% of GDP | 5.3% of GDP | ₹54.8 |
RBI reports 2023–24 gross financial assets of ₹34.32 lakh crore and liabilities of ₹18.79 lakh crore. Net financial assets were ₹15.52 lakh crore. The ratio above is a KalQLater calculation, not an RBI headline.
Deposits remain the anchor; the rest of the mix is broader
| Instrument | ₹ crore | Share of gross flow | Interpretation |
|---|---|---|---|
| Deposits | 13,91,244 | 40.5% | Largest destination; total includes an RBI reclassification effect |
| Provident & pension funds | 7,18,661 | 20.9% | Mandatory and voluntary retirement saving |
| Life insurance funds | 5,90,999 | 17.2% | Insurance-linked financial acquisition |
| Small savings | 3,09,751 | 9.0% | Government small-saving instruments |
| Investments | 3,02,834 | 8.8% | Includes mutual funds, equity and other investments |
| Currency | 1,18,026 | 3.4% | Physical currency acquisition |
Within “investments,” mutual-fund flows were ₹2.39 lakh crore and equity flows ₹29,080 crore. They should not be added again to the parent investments total.
Up from ₹15.63tn in August 2016. This growth combines investor flows, market returns and institutional money.
Folios are accounts, not unique people. One investor may hold several folios.
AMFI counted 10.75 crore outstanding SIP accounts; accounts are not households.
These measures show distribution and market deepening, not a population-wide migration from deposits into mutual funds. The RBI flow table still shows deposits as the largest single bucket.
Bigger loan books do not automatically mean more indebted households
AIDIS household incidence
| Residence | Any debt | Institutional only | Non-institutional only | Both |
|---|---|---|---|---|
| Rural | 35.0% | 17.8% | 10.2% | 7.0% |
| Urban | 22.4% | 14.5% | 4.9% | 3.0% |
Assets were still property-heavy
| Residence | Land | Buildings | Deposits | Other |
|---|---|---|---|---|
| Rural | 69% | 22% | 5% | 4% |
| Urban | 49% | 38% | 9% | 4% |
AIDIS is periodic, uses a survey reference date and includes formal and informal debt. RBI’s bank-credit series is monthly, covers lenders in its reporting system and measures rupees outstanding. Those are different questions.
What banks were lending for in March 2024
| Category | Outstanding | Share of personal loans | Year-on-year growth |
|---|---|---|---|
| Housing | 27,18,715 | 51.0% | about 12.1% excluding merger effect |
| Other personal loans | 14,02,775 | 26.3% | about 6.6% excluding merger effect |
| Vehicle loans | 5,73,398 | 10.8% | 7.3% |
| Credit-card outstanding | 2,57,016 | 4.8% | 13.6% |
| Education | 1,19,380 | 2.2% | 13.8% |
| Consumer durables | 23,713 | 0.4% | −0.9% |
Housing was about half of the personal-loan book. Classification changes and the 2023 bank–non-bank merger make long-run category-growth comparisons unsafe unless adjusted, so this page treats March 2024 as a dated composition snapshot.
The basket moved beyond staples, not beyond food
| Survey year | Rural food | Rural cereals | Urban food | Urban cereals |
|---|---|---|---|---|
| 2004–05 | 55.0% | 18.0% | 42.5% | 10.0% |
| 2011–12 | 52.9% | 10.75% | 42.62% | 6.66% |
| 2022–23 | 46.38% | 4.91% | 39.17% | 3.64% |
| 2023–24 | 47.04% | 4.99% | 39.68% | 3.76% |
In 2023–24, non-food items represented about 53% of rural MPCE and 60% of urban MPCE. Among food groups, beverages, refreshments and processed food had the largest share; among major non-food groups, conveyance, clothing and footwear, miscellaneous goods and entertainment, and durable goods were prominent.
The small increase in food share from 2022–23 to 2023–24 does not reverse the two-decade structural fall. Nor does a falling cereal share mean cereal consumption disappeared: it means cereals occupied a smaller percentage of total measured spending.
What changed—and what did not
- Liabilities became much larger relative to annual financial-asset acquisition.
- Formal market participation expanded through mutual funds and recurring SIPs.
- Food—and especially cereals—took a smaller share of household spending.
- Housing dominates formal personal credit, while unsecured categories also grew.
- Deposits remain the largest annual financial-asset destination.
- Property dominates household asset value in the latest AIDIS benchmark.
- Rural debt incidence remains higher than urban incidence in that survey.
- No single official dataset follows borrowing, portfolios and spending annually for 20 years.
The defensible conclusion is a change in composition and balance-sheet intensity—not that every Indian household became a market investor or an overextended borrower.
Popular narratives need narrower wording
| Claim | What the data supports | What it cannot establish |
|---|---|---|
| India moved from save-first to borrow-invest-spend | Borrowing flows rose relative to new financial assets; market participation and non-food budget share expanded. | A single behavioural model for every household or proof of individual motivation. |
| Mutual funds replaced deposits | Mutual-fund AUM, folios and SIP accounts expanded strongly. | Deposits remain the largest new financial-asset bucket; AUM is not pure contribution. |
| Personal loans drove household debt | Formal personal credit expanded and housing is about half of the selected-bank personal-loan book. | All liabilities, informal loans or the causal contribution of each purpose. |
| Indians stopped saving | Net household financial saving was lower relative to GDP than in some earlier points. | That households stopped acquiring assets or that financial saving equals total saving. |
| Consumer spending shifted beyond food | Food and cereal shares of HCES MPCE fell over two decades. | That nominal food spending fell, or that PFCE and HCES categories can be spliced. |
For wider monetary context, see Is the US Dollar Actually Dying?.
Seven datasets, seven different questions
| Dataset | Measures | Population / coverage | Frequency | Best use | Main limitation |
|---|---|---|---|---|---|
| RBI household financial flows | New assets, liabilities and net financial saving | Household sector in national accounts | Annual | Macro flow ratios | Not the same households over time; revisions |
| RBI financial stocks | Balance-sheet asset and liability stocks | Experimental household-sector estimates | Annual / experimental | Debt-stock context | Shorter series and valuation effects |
| RBI personal-loan credit | Outstanding bank credit by purpose | Reporting scheduled commercial banks | Monthly | Formal loan composition | Excludes informal debt; classifications change |
| AIDIS | Debt incidence, source, amount and assets | Sampled rural and urban households | Periodic | Who reports debt and from whom | Not annual; survey reference dates |
| HCES | Household consumption expenditure | Sampled rural and urban households | Survey rounds | Budget shares | Questionnaire and recall-method changes |
| PFCE | Economy-wide private final consumption | National accounts | Annual / quarterly | Macroeconomic consumption | Not survey household budgets; base-year changes |
| AMFI | AUM, folios and SIP accounts / contributions | Indian mutual-fund industry | Monthly | Market participation infrastructure | AUM includes returns; accounts are not people |
How the reconstruction was built
- Use RBI household financial accounts for annual flows and keep preliminary/revised labels attached.
- Calculate liabilities per ₹100 of gross financial assets only where both values use a comparable GDP denominator.
- Use AIDIS for household incidence, lender type and asset composition—not as an annual debt-total series.
- Use HCES for household expenditure and consumer spending shares, retaining rural/urban splits. Shares are not nominal spending levels.
- Use RBI bank-credit tables for a dated formal-loan composition snapshot, with sample and merger cautions.
- Use AMFI administration data for industry AUM, folios and SIPs; do not equate accounts with people.
- Display N/A for incompatible or unverified years. No interpolation, smoothing or inflation adjustment is used for percentage shares.
Download the data used in this analysis
The CSV identifies the section, measure, reference year, geography, value, unit, source agency and comparability note. Retrieval date: 4 October 2026. Derived ratios use financial liabilities ÷ gross financial assets × 100; allocation shares use instrument flow ÷ gross financial-asset flow × 100. Blank values preserve unavailable comparisons.
Limits
Financial flows can be revised. “Household” can include non-profit institutions in some national-account tables. AUM reflects market prices as well as contributions. Surveys have sampling error, recall periods and questionnaire changes. National-account base years and category definitions change. Bank credit excludes informal lending and other liabilities. AIDIS is not annual. PFCE is macroeconomic consumption, while HCES is survey-based household expenditure. Nominal rupee values are affected by inflation; this page compares shares and ratios and therefore applies no inflation deflator. The newest observation differs by source: 2023–24 for RBI flows and HCES, 2018 reference date for AIDIS, March 2024 for the credit composition shown, and August 2026 for AMFI. The 2019–20 and 2020–21 period was disrupted by COVID, so pandemic spending is not used as a normal structural benchmark. Aggregate data cannot identify individual motivations.
Sources
- RBIHousehold Financial Savings, 2023–2421 May 2025 · flows for 2021–22 to 2023–24
- RBIHousehold balance sheets in IndiaFinancial assets and liabilities stock to March 2023
- MoSPI / NSSHCES 2023–24 final reportSurvey spending composition
- MoSPI / NSSHCES 2004–05 reportHistorical consumption benchmark
- MoSPI / NSSAIDIS Report 588Debt incidence and asset composition; reference date 30 June 2018
- RBISectoral Deployment of Bank Credit — March 2024Select-bank credit; 95% of non-food credit
- AMFIIndian mutual-fund industry dataAUM and folios; August 2026
- AMFISIP dataMonthly SIP contribution and accounts; August 2026
Browse more Money & Consumer research, read the site’s methodology, or explore the full research index.