Money, Work and the Economy

The complete report has seven parts and one 204-page master PDF. The research cutoff for this part is 2 October 2026. Every figure retains its own observation year. Page references in the original reading guide below refer to this volume’s printed pages within the master PDF; use the contents here to move through the web edition.

Why I wanted this report

An author's note

I wanted to sit down with one question: what actually changed in India after 2014? It sounds simple until you start reading the numbers. One chart talks about roads. Another talks about jobs. Both may be telling you something real, while leaving out the part you need to understand it.

I write at MisterLove because I like taking a difficult subject apart until it becomes clear. This report follows that approach. I want someone who normally closes a long article after two paragraphs to be able to follow the whole thing. You should not need an economics degree to understand what is happening around you.

The comparisons shared by The Matrix India gave this project a starting point. I have treated those charts as questions to investigate. Wherever possible, the numbers here lead back to the organisation that collected or published them. A figure can be correct and still give the wrong impression when its date, definition or starting point is missing.

I have kept the improvements, the problems and the awkward gaps. If a survey began after 2014, I say so. If a figure is a forecast or a budget plan, I name it as one. If two numbers cannot fairly be compared, I explain why.

This first volume explains income, work, prices and public money. The seventh part gives my final comparison of the two governments and states the priorities behind it. I want you to have enough information to understand that judgment and form your own view.

For readers meeting me here: I'm Lovepreet Singh, a researcher and writer, founder of Lovelace, and an MBA student in Business Strategy and Analysis. My work at MisterLove brings together questions about society, history, public policy and technology. Clear writing is the thread running through it.

Lovepreet Singh
MisterLove · 2 October 2026

Sources: 1, 32, 35, 36.

Seven parts. One country.

This volume starts with money and work.

The six evidence volumes follow the same approach: show the earlier number, show the recent number, explain what changed, and make the limits visible. The seventh part draws the final comparison between the governments.

All seven parts are complete and readable here. The master PDF contains the original 201 pages, plus a cover, an author’s note and linked contents. It is 204 pages in total. Each volume includes its own sources.

Inside Part 1

This part covers the terms used in the comparisons, the economy, income, prices, wages, work, women’s participation, household spending, poverty, public money, banking, trade, debt, investment and the starting economic chart. Start at the beginning if the terms are new, or use the contents to find one subject.

What does “before and after” mean?

The dates are part of the number.

For a yearly figure, 2013-14 is usually the nearest full financial year before the change of Union government in May 2014. A financial year runs from 1 April to 31 March. So 2025-26 means April 2025 through March 2026. It is already a completed year at this report's cut-off.

A survey labelled 2023-24 may follow a different calendar. The consumption survey ran from August 2023 to July 2024. The older labour survey used July to June. Its 2025 edition uses January to December. The period is written wherever it matters.

One lakh is 100,000. One crore is 10 million. One lakh crore is one trillion rupees. A billion dollars is 100 crore dollars; it is not a billion rupees. The currency matters.

“Latest” here means a release checked by 2 October 2026. It does not mean every measure describes October 2026. Some reliable surveys take years to appear. Blank evidence is left blank; a future target is never counted as an achieved result.

Sources: 5, 9, 12, 18, 19.

A bigger economy. How much bigger?

GDP counts production across the country.

GDP, or gross domestic product, adds up the value of final goods and services produced inside India. A finished shirt counts; adding the full price of its cotton again would count the same production twice. Shops, farms, factories, software and many other activities contribute to the total.

In that historical series, the money value of annual output was about 2.94 times the earlier level. That does not mean India produced 2.94 times as much. Part of the increase reflects higher prices. Nor does it mean each household received 2.94 times as much income.

The newer 2026 reading

MoSPI introduced a new GDP series with base year 2022-23 and updated it again in August 2026. In the latest released quarter, April-June 2026, nominal GDP was ₹88.27 lakh crore and real GDP growth was 7.8% over April-June 2025. Both sides of that quarterly calculation use the new series.

The historical table above deliberately retains the older series for its long comparison. Its 2024-25 number is no longer the latest revised GDP level. The 2026 quarterly figure is shown separately because a quarter is not a full year and the statistical method changed. Dividing the two would produce a misleading answer.

Production tells us what the country can generate. The next question is how much of that reaches a person, and how much that person's money can buy.

Historical series and current quarter are separate comparisons; no cross-series growth rate is calculated.

Sources: 2, 3, 5, 6.

Why lower inflation still feels expensive

A slower price rise is still a price rise.

The first number is a whole-year average. The later two compare one month with the same month a year earlier. They are useful dated readings, but they are not a matched pair of annual averages. August 2026 uses the new CPI base of 2024=100 and is provisional.

Imagine the same basket costs ₹100, then rises to ₹110. Inflation was 10%. If it next rises to ₹115.50, inflation has fallen to 5%. The basket has still become dearer. That is why “inflation is lower” can be true while a family is still struggling with the bill.

CPI means Consumer Price Index. It follows a weighted basket of household purchases. Food, rent, transport and other items do not receive equal weight. A family spending most of its income on food can feel a different squeeze from one spending more on travel or services.

The new basket is meant to reflect newer spending patterns. This also means a long comparison needs the official treatment of the series change. Simply taking an old index number and dividing it by a new one will not tell you the rise in prices since 2014.

The Matrix chart's 2.1% is not the latest monthly CPI reading at this report's cut-off. Whatever period a chart uses, it needs to name that period. For your own budget, compare the prices you pay with the growth in your take-home income. That is where the pressure shows up.

Sources: 7, 8, 32.

The average Indian is not an average

A national average is a useful starting point, not a payslip.

These figures come from the same historical national-accounts series in the January 2026 Economic Survey. They are annual amounts per person. The later figure was provisional. This table is a consistent older-series comparison, not a claim to show the latest rebased income estimate.

The nominal figure rose by about 160%. The real figure rose by about 67%. The smaller real increase removes the price changes built into the national accounts. Over eleven years, that real increase works out to roughly 4.8% a year when compounded. These percentages are calculations from the table.

Net national income allows for the wearing out of productive assets and includes residents' income from abroad, net of income going abroad. Dividing it by the population gives per capita income. Children and people without earnings are in that population too.

Now imagine five people earning ₹10,000, ₹10,000, ₹10,000, ₹10,000 and ₹1,00,000 a month. Their average is ₹28,000. Four of the five receive much less. Their median, the middle person's income, is ₹10,000. This is an illustration, not an estimate of Indian incomes.

A rise in average income matters. It tells us the economy generated more income per person on this measure. To understand everyday life, we also need wages, job stability, consumption and the distribution of the gains. None of those can be read straight from the average.

Sources: 4, 5.

What did workers actually earn?

Monthly salary, business earnings and daily wages are different measures.

These PLFS annual averages use workers' current weekly status. Salary covers the preceding calendar month; self-employment earnings cover the last 30 days. Casual pay is daily and excludes public works. Amounts are nominal. Sampling changed in 2025, so the years are not strictly comparable. This is a recent picture, not a wage series from 2014.

The rows show why a single statement about “wages” can be too broad. The reported male salary average increased; the male casual daily average was almost unchanged. Workers can face very different outcomes inside the same growing economy.

These averages do not follow the same individual over time. A change in who enters a job category can move the average. Nor is a daily wage a monthly income: someone paid ₹455 for ten days receives much less than someone finding work on twenty-six days.

To test buying power, pay must be adjusted for a price index covering the same period. Comparing a 2025 annual wage rise with August 2026 inflation would be a date mismatch. For example, a 6% pay rise with 5% inflation gives a real rise of about 0.95%, calculated as 1.06 divided by 1.05, minus one.

The useful questions are therefore about the pay rate, the number of days worked, prices, and how predictable the next payment is.

Sources: 9, 10.

What counts as having a job?

The headline unemployment rate leaves out part of the story.

These PLFS annual periods run from July to June. The starting point is 2017-18 because that is when this survey series began. It must not be relabelled “2014”. A longer employment history needs the older surveys and a careful explanation of differences.

The labour force means people working or seeking and available for work. Its share of the population is called LFPR. The worker population ratio, or WPR, counts only the people working. The unemployment rate asks what share of the labour force is without work. Its denominator is smaller than the whole population.

Consider 100 adults. If 60 are in the labour force and three of those are unemployed, the unemployment rate is 5%, not 3%. The other 40 can include students, people doing household work and others outside the labour force. They are not all recorded as unemployed.

A worker may have a monthly salary, run a shop, do casual labour or help without a separate wage in a family enterprise. “Employed” does not promise good pay, full-time hours or a secure contract. The usual-status measure looks mainly across the preceding year; weekly measures capture a shorter window.

The old survey series records both higher participation and lower unemployment over the period shown. That is useful evidence. The next step is to examine the type of work and earnings, rather than treating every counted worker as someone with a comfortable salaried job.

Sources: 11, 10.

Work is widespread. Secure work is another question.

The complete 2025 annual survey is the recent reference here.

The first four rows use different populations from the last three. The job-type rows divide employed people into groups and add to 100%. The youth unemployment figure is for young people in the labour force; it does not mean 9.9% of all young people are unemployed.

The survey's sample design changed from January 2025. The ministry explicitly says the results are not strictly comparable with earlier annual PLFS reports. A line drawn through 2017, 2024 and 2025 without marking that break would look more certain than the evidence permits.

Self-employment can mean a successful business, a small farm or work taken up because better options are scarce. Regular salary also does not automatically mean a written contract, paid leave or pension cover. The categories tell us the form of work; they do not measure its whole quality.

For a young person trying to begin a career, the youth figure is often more relevant than the all-age average. For a household trying to pay rent every month, steady earnings and reliable hours can matter more than the simple fact of being counted as employed.

Sources: 9, 10.

Participation and independence are different questions

More recorded work needs to be read alongside pay and control over money.

Within the older survey design, women's participation increased by 18.4 percentage points. That is a substantial rise in the share of women recorded as working or looking for work. The 2025 row belongs to the revised survey and should not be read as a clean 1.7-point fall from 2023-24.

A woman helping on a family farm or in a family shop can count as employed without receiving a separate salary. That work contributes to production, but it does not automatically give her an independent income she controls. Unpaid household care is also economically valuable, although it is treated differently in these employment statistics.

The 2025 survey places 64.2% of employed women in self-employment and 18.2% in regular wage or salaried work. Its reported average monthly salary for women is ₹18,353, compared with ₹24,217 for men. These broad averages are not a comparison of men and women doing exactly the same job for the same hours.

Differences in occupation, hours, seniority and location all matter. So can barriers such as childcare, safety, transport and access to suitable work. The figures alone cannot say how much of the pay gap each factor causes.

The practical question is whether a woman can choose worthwhile work, receive a fair income and decide how to use it. Participation is one part of that picture. Pay, time and freedom of choice complete much more of it.

Sources: 9, 10, 11.

How much does daily life consume?

Spending per person brings the discussion closer to the kitchen table.

These HCES figures exclude the imputed value of specified free items received through welfare programmes. Imputation means adding an estimated money value for something received without payment. Mixing a series that includes those values with one that excludes them changes the comparison.

At current prices, the rural amount is about 2.88 times its earlier level and the urban amount about 2.66 times. After the published price adjustment, the increases are about 45% and 38%. These are changes over roughly twelve years, not one year.

Monthly per capita consumption expenditure, or MPCE, divides household consumption by household members. A five-person household with MPCE of ₹4,122 has about ₹20,610 of measured monthly household consumption. That multiplication is an illustration. It does not say every rural family spends that amount or has five members.

The surveys changed their questionnaires, coverage of items and interviewing approach between 2011-12 and 2023-24. The table is the published comparison, with those limits. It should not be treated as a perfectly unchanged measuring instrument.

Consumption is not the same as salary. A household can use savings, borrowing, transfers or goods it produces itself. Higher measured consumption is relevant to living standards, but it does not by itself tell us whether the household is financially secure. The next page asks how many people remain below specific spending thresholds.

Sources: 12, 13.

Fewer people below the line. Which line?

A poverty rate only makes sense when the threshold is visible.

The April 2026 India Development Update reports these estimates using consumption surveys. The $3 line is the international extreme-poverty threshold. The $4.20 line is commonly used for lower-middle-income economies. Raising the threshold captures a larger group with limited resources.

PPP means purchasing power parity. It adjusts for differences in what money buys across countries. These are international dollars at 2021 purchasing power. Multiplying $3 by today's dollar-rupee exchange rate would give the wrong local threshold.

On these measures, the reduction is large: 24.5 percentage points at the extreme-poverty line and 41.6 points at the higher line. The World Bank estimates about 35 million people were still below the extreme-poverty line in 2023-24. That is about 3.5 crore people.

Moving just above a line does not make a person comfortably middle class. A family may still be one illness, crop loss or lost job away from difficulty. The threshold is a way to count a defined deprivation; it is not a complete description of a decent life.

These estimates also span 2011-12 to 2023-24, including years before 2014. They cannot assign the whole fall to one administration. Changes in survey methods and the construction of comparable consumption measures affect the estimate. Monetary poverty, multidimensional poverty and hunger measure different things. Part 3 will examine those differences in detail.

Sources: 14, 13.

Poverty and inequality can move in different directions

Income is money received. Wealth is what is owned, minus what is owed.

Suppose one person earns ₹10,000 and another ₹1,00,000. Later they earn ₹15,000 and ₹2,00,000. Both have more money. The poorer person's position has improved, yet the income gap has widened. Falling poverty and rising concentration at the top can therefore happen together.

These are research estimates, not a direct government census of every person's income and assets. The researchers combine surveys, tax information, national accounts and information on very wealthy people. The World Inequality Report 2026 also discusses India's high concentration of income and wealth.

The date matters: a report published in 2026 does not automatically turn every underlying observation into 2026 data. The table here deliberately keeps the 2022-23 observation year attached to the values. It does not invent a matched 2014 observation from a different source.

Consumption surveys answer another question: how much households consume. A wealthy household can save a large part of its income. Its spending share can therefore be much smaller than its income or wealth share. A consumption-inequality measure cannot simply replace an income-inequality measure.

The exact concentration estimates depend on how missing or under-recorded income at the top is handled. That deserves scrutiny. It also deserves to be read alongside the evidence of falling poverty, rather than assuming one cancels the other.

For everyday life, the issue is access: who can absorb an emergency, buy a home, fund education or take a business risk? Average income alone leaves those differences almost completely hidden.

Sources: 15, 16, 14.

The gap between receipts and spending

The fiscal deficit measures a year's financing gap.

The fiscal deficit is total government spending minus its receipts other than new borrowing. It is a flow over a year. GDP is the whole economy's annual output, so 4.4% of GDP does not mean 4.4% of the government's budget.

CGA's provisional 2025-26 accounts show ₹49.05 lakh crore of spending and ₹33.86 lakh crore of non-debt receipts. The gap is about ₹15.19 lakh crore. Figures are rounded here from the accounts; the source reports a 4.40% deficit ratio. These are unaudited provisional actuals, not the earlier Budget Estimate.

The 2026-27 figure is a plan for a year still in progress. The historical and current ratios also reflect the GDP estimates used by their publishers. GDP revisions can change a ratio even when the rupee borrowing amount stays the same. Do not read the tiny difference between 4.5 and 4.4 as a perfectly precise measure of performance.

Borrowing can help pay for useful long-lived assets or support activity in a downturn. It also creates future interest costs. The amount, purpose, financing cost and future income all matter.

The two endpoints hide the years in between, including the pandemic. A fair assessment needs that path as well as the endpoints. These figures cover the Centre. State finances must be added with care to assess government as a whole, without double-counting transfers between governments.

Sources: 17, 18, 19.

The old bills travel with the new budget

Debt is the outstanding stock. A deficit adds to the financing need.

These ratios describe central government debt in the Budget's framework. They are not the combined debt of the Centre and every state, and they are not India's external debt. The GDP assumptions are those of the February 2026 Budget, before the later national-accounts rebasing.

The Matrix chart uses 51.6% for 2013-14 and 56.1% later. I have not established a common debt definition and GDP vintage for that pair. I therefore do not use it to calculate a precise change. Historical liabilities figures can differ because of exchange-rate valuation and which liabilities are included.

Debt is what remains to be repaid at a point in time. The fiscal deficit is a year's gap. A government can reduce its annual deficit and still add to its debt. A debt-to-GDP ratio can fall while the rupee debt rises, if nominal GDP grows faster.

The provisional 2025-26 accounts put central interest payments at ₹12.43 lakh crore. Against revenue receipts of ₹33.02 lakh crore, that is about 37.6%, calculated from CGA's figures. In plain language, a large part of recurring receipts is already committed to interest before other uses are considered.

A government is not identical to a household: it can tax, refinance debt and shape economic activity. It still faces choices. More money committed to interest can leave less room for other priorities unless revenue grows, borrowing rises or spending changes.

The useful question is whether debt can be serviced while maintaining essential services and productive investment. One debt ratio, read alone, cannot answer it.

Sources: 19, 18, 32.

More spending on assets. Then comes delivery.

Capital expenditure is a budget category, not a completed-project count.

The first amount is ₹1,87,675 crore in the Budget's historical account. The second is ₹10,69,119 crore in CGA's provisional accounts. The cash amount is about 5.70 times the earlier level. That is a nominal increase across twelve years, before adjusting for construction and other costs.

Capital spending includes acquiring or creating assets and certain loans or investments. Roads, railway works and equipment are familiar examples. The Centre can also give grants that create assets elsewhere; those grants are not all booked under its own capital-expenditure heading.

For 2025-26, direct capital spending was about 21.8% of total central expenditure, calculated as ₹10.69 lakh crore divided by ₹49.05 lakh crore. The Matrix chart's 22.1% corresponds closely to revised budget amounts. Actual spending and revised plans should keep their labels.

It is tempting to call every capital rupee good and every recurring rupee wasteful. That misses how services work. A new hospital needs doctors, nurses, medicines and maintenance. A school needs teachers. Many useful costs recur each year.

The more demanding test is what the spending delivered: a functioning asset, the people using it, the time or cost saved, and the upkeep it needs. Announced allocations, money spent and completed projects are three different stages. Part 2 will follow that distinction through physical infrastructure.

Sources: 21, 18, 20, 32.

Where the money comes from, and who it helps

Gross tax, net tax and subsidies answer different questions.

Gross central tax receipts are counted before the states' share is transferred. Net tax revenue is what remains for the Centre after the relevant deductions and transfers. Placing gross receipts in one year beside net receipts in another produces a false comparison.

The 2026-27 gross number is a budget plan, not money already collected. A larger cash total over a long period can reflect a bigger economy, rising prices, tax changes and better collection. It does not by itself measure the tax burden on a typical household.

CGA's 2025-26 major-subsidy total includes about ₹2.28 lakh crore for food, ₹2.11 lakh crore for fertilisers and ₹0.14 lakh crore for petroleum. The displayed components are rounded. This heading is narrower than all welfare spending: pensions, health programmes and many other forms of support appear elsewhere.

A subsidy can reduce a price or support access to a basic need. Its effect depends on who receives it, whether it reaches them and what it costs to deliver. Spending less is not automatically better; spending more is not automatically more effective.

The same care applies to GST. A total GST collection series begins after the 2017 introduction of the system. Writing “zero GST in 2014” would not mean there were zero indirect taxes then. A serious comparison must include the taxes that GST replaced.

Sources: 17, 18, 20.

Bad loans became a smaller share of lending

The improvement matters. So does the route taken.

An NPA is a loan classified as non-performing under banking rules, generally because required payments have remained overdue beyond the applicable limit. The gross NPA ratio compares these bad loans with total gross advances. “Gross” is before deducting provisions, the amounts banks set aside for possible losses.

The later ratio is lower by 2.3 percentage points. In simple terms, a smaller share of the recorded loan book was classified as bad at that date. That supports a healthier banking system, but it is not a measure of how many borrowers are debt-free.

The path was not a steady fall from 2014. Recognition of stressed loans changed during the clean-up, including RBI's Asset Quality Review in 2015. Banks can reduce NPAs through repayment, recovery, restructuring that meets the rules, sales and write-offs. A growing total loan book can also lower the ratio.

A write-off changes the bank's accounting. It does not automatically cancel the borrower's obligation or mean that the full amount was recovered. Those outcomes need separate evidence. Likewise, a low bad-loan ratio today does not guarantee that every new loan will be repaid.

RBI's June 2026 report also records strong lending growth and examines banks' ability to handle stress. Its stress scenarios are tests of possible future conditions, not forecasts that a crisis will occur. For a borrower, the further questions are whether credit is accessible, affordable and suited to the income that must repay it.

Sources: 22, 23, 24.

A bank account became much more common

Access is a real gain. Use and control are the next questions.

These Global Findex figures count ownership of an account at a financial institution or with a mobile-money provider. They are survey estimates for adults, not the number of accounts. One person can own several accounts, so an administrative account total cannot simply be divided by the population to measure access.

Overall ownership rose by 35.9 percentage points between the two survey years. The large gap between women and men in 2014 had almost disappeared in the 2024 estimates. Small differences such as the later 0.4-point female lead should not be overinterpreted without considering sampling uncertainty.

An account can make it easier to receive wages or transfers, store money and make payments. But an account with no balance, one controlled by somebody else, or one that is difficult to access is a different experience from a regularly used personal account.

To judge financial inclusion more fully, look at activity, independent control, savings, affordable borrowing, service costs and the ability to resolve problems. Having the doorway is valuable; it does not tell us everything about using the room.

The later figures were collected in 2024 and published in the 2025 Findex edition. They are not observations from 2026. The 2014 survey is also a calendar-year observation, not a clean measurement taken just before the May change of government. This is a strong decade-long comparison of access, with those dates kept visible.

Sources: 25.

Exports grew. Goods and services both matter.

Keep the basket and currency the same.

The later combined total is about 85% above the earlier one in dollar terms. Spread over twelve years, that works out to roughly 5.3% a year when compounded. These are calculations using the displayed, partly rounded values; they are not inflation-adjusted export-volume growth.

The ministry's later 2025-26 estimate separates roughly $442 billion of goods from $421 billion of services. Goods include physical products. Services can include software, business services and other activities sold to customers abroad. Comparing old goods-only exports with a new goods-plus-services total exaggerates the change.

The original April 2026 combined estimate was about $860.09 billion. Services data were then partly estimated and were updated later. This volume uses the later rounded $863 billion figure and retains its estimate label. Extra decimal places would imply precision that the selected release does not provide.

The latest partial-year release reports combined exports of $399.27 billion during April-August 2026, against $345.55 billion in the same months of 2025. That is a like-period increase of 15.55%. Five months must be compared with five months, not with a whole year's exports.

Export earnings can support firms, jobs and foreign-currency receipts. Gross exports are not the same as domestic value added: producing an exported item may require imported parts, fuel or services. Imports also need to be considered before drawing a conclusion about the country's external balance.

US dollars throughout. Nominal values; services estimates may be revised.

Sources: 26, 27, 28.

More foreign debt can coexist with a lower debt ratio

External debt covers money owed to non-residents by several sectors.

The September 2026 RBI release provides these historical and recent figures in the same table. The dollar debt stock increased by about 74%. Its share of GDP was lower. Those statements can both be true because the economy is larger. The chart's earlier 23.3% is not the March 2014 ratio in this RBI table.

External debt is not just the Union government's borrowing. Companies, banks and other borrowers are included. Nor must every external loan be denominated in dollars: “external” refers to the creditor being a non-resident.

Reserves are foreign-currency assets and other reserve assets held by the monetary authorities. They provide a buffer, but they are not an unlimited fund for meeting every private debt. The reserve-to-debt ratio is also different from import cover, which asks how many months of imports reserves could pay for at a specified import rate.

A current account deficit is a flow, not this debt stock. It combines trade in goods and services with income and transfers across borders. The 2013-14 deficit was reported as 1.7% of GDP. This volume does not certify the chart's later 0.6% without a matched balance-of-payments release and revision date.

The timing of repayment matters as much as the total. Debt falling due soon, currency changes and the ability to refinance can create pressure. A single external-debt ratio cannot capture all of that risk.

Sources: 29, 33, 32.

Money entering the country is only one side

Gross FDI is different from net FDI.

These are total FDI inflows, not just equity inflows. DPIIT's March 2026 fact sheet includes equity, estimated reinvested earnings and other capital. The latest annual total is about 2.62 times the 2013-14 level, an increase of roughly 162% in nominal dollar terms.

FDI, or foreign direct investment, generally involves a lasting ownership interest in an enterprise. It can bring money, technology, management links and access to markets. Its effects depend on the investment's form and what the business actually does.

Gross inflow counts the inward side. Net measures also deduct specified outward flows, such as investor exits or repatriation; broader net direct-investment balances also account for residents' investment abroad. Always check the definition in the source. A strong gross total can coexist with a weak net figure.

Buying an existing business and building a new factory are both capable of attracting FDI. They do not create the same amount of new productive capacity at the moment of the transaction. Reinvested profits can count even when fresh money does not cross the border that month.

The fact sheet marks components as estimates and the recent data as provisional. The total is useful evidence of the scale of inward investment. It is not a direct count of factories built, jobs created or money permanently staying in India. Those outcomes need their own evidence.

Sources: 30, 31.

Where the economic comparison needs care

First eight rows of the fifteen-indicator chart.

The chart is a starting list of claims. The notes below explain what this volume established, changed or left unresolved. A “before” and “after” pair needs the same concept, period and source treatment.

The original 4.7% figure is a historical vintage, not a fabricated number. The issue is comparability.

Sources: 6, 7, 8, 17, 18, 19, 20, 22, 23, 32, 33, 34.

The remaining rows, and the question of credit

A change after 2014 does not, by itself, identify its cause.

Several chart rows remain unresolved here; they are not promoted into verified facts. Part 6 extends the chart and ranking audit. The two posts originally supplied, covering infrastructure and Scheduled Tribe crime, belong to Parts 2 and 5 respectively.

For responsibility, timing is a starting clue. A result can reflect Union policy, state policy, earlier investment, private effort, technology and world conditions. Establishing how much one policy caused requires evidence beyond two endpoints. Recognising that does not erase an achievement or a failure; it makes the explanation more accurate.

Sources: 18, 21, 23, 29, 32, 35, 36.

Sources and reading notes