Saving, investment and capital productivity: Harrod-Domar

Detail

1. Capital-output ratio (K/Y)

• Capital-output ratio (K/Y) is the amount of capital (machines, buildings, roads, equipment) needed to produce one unit of output.

• Example: a factory worth ₹400 crore makes goods worth ₹100 crore a year. K/Y = 400 ÷ 100 = 4.

• A lower ratio means capital is used more productively. The same output comes from less capital.

• K/Y shows the average productivity of all capital. The capital stock was built up over many years.

2. Incremental capital-output ratio (ICOR)

• ICOR = ΔK / ΔY. It is the extra capital needed to produce one extra unit of output. (Δ means "change in".)
• Why ICOR can be measured from the national accounts:
• The change in capital stock (ΔK) is the year's investment (I).
• Divide the top and bottom of ΔK/ΔY by GDP (Y).

• So ICOR ≈ investment rate (I/Y) ÷ GDP growth rate (ΔY/Y).

• Lower ICOR = more efficient investment. Each rupee invested adds more output.

• Worked example (scaffold): investment of 32% of GDP with 8% growth gives ICOR = 32 ÷ 8 = 4.
• Worked example (reverse use, for planning): suppose India wants 8% growth.
• With ICOR = 4, it needs investment of 8 × 4 = 32% of GDP.
• With ICOR = 4.5, it needs 8 × 4.5 = 36% of GDP.

• Lesson: a small rise in ICOR (waste, delays, idle capacity) sharply raises the saving the country must find.

• What raises ICOR:

• Long project delays and cost overruns.
• Heavy, long-gestation infrastructure (ports, power plants), which pays off slowly.

• Idle capacity during a slowdown. Capital exists but output does not rise.

• What lowers ICOR: better technology, skilled labour, faster project completion and fuller use of existing capacity.

3. Harrod-Domar model (Harrod 1939, Domar 1946)

• The model was built separately by Roy Harrod (1939) and Evsey Domar (1946). It grew out of Keynesian thinking after the Great Depression, which showed that an economy can get stuck below full employment (Class 12, Introduction).
• Formula: g = s / v
• g = growth rate of output (GDP).
• s = saving rate (saving ÷ GDP). The model assumes it equals the investment rate.

• v = ICOR (capital-output ratio).

• Worked example (scaffold): saving of 30% and ICOR of 4 give 30 ÷ 4 = 7.5% growth.

• Short derivation (for understanding):
• Saving = s × Y, and saving = investment, so I = sY.
• Investment adds to capital, so ΔK = sY.
• Output rises by ΔY = ΔK ÷ v = sY ÷ v.

• Growth rate = ΔY ÷ Y = s ÷ v.

• How it works (the chain):

• Saving funds investment. Households and firms save, and banks lend those savings to investors.
• Investment adds to the capital stock.

• Given the fixed ICOR, more capital means more output.

• Domar's point: investment has a "dual character".

• Demand side: new investment raises income through the multiplier. This links to Class 12, Determination of Income and Employment: ΔY = ΔI × 1/(1 − MPC), where MPC (marginal propensity to consume) is the share of extra income that people spend.
• Supply side: the same investment also adds to productive capacity.
• Steady growth needs demand to grow exactly as fast as capacity. If it doesn't, factories stand idle (deficient demand) or demand runs ahead of supply (excess demand, which pushes prices up).

4. Three growth rates and the "knife-edge"

• Actual growth rate (G): the growth that really happens (G = s / actual ICOR).
• Warranted growth rate (Gw): the rate at which firms are content, because the capacity they build is fully used. Gw = s ÷ the capital-output ratio firms want.
• Natural growth rate (Gn): the highest rate allowed by labour-force growth plus technical progress. It is the ceiling for long-run growth.

• Example: labour force grows 1.5% a year and technical progress adds 2%, so Gn ≈ 3.5%.

• "Knife-edge" instability: steady growth needs Gw = Gn = actual growth. This happens only by chance, because s, v and labour growth are set by different people for different reasons.

• Any gap tends to widen, not correct itself.
• If actual growth is above Gw → firms find their capacity too small → they invest more → the multiplier raises demand further → growth moves even further above Gw.

• If actual growth is below Gw → factories stand idle → firms cut investment → demand falls further → a slump.

• The result is either chronic unemployment or inflation.

• Gw > Gn: the economy saves more than its labour force can use → capital lies idle → long stagnation and unemployment.

• Gw < Gn: labour grows faster than capital → there are not enough machines to employ everyone (structural unemployment), while demand runs ahead of capacity (inflationary pressure).

• This instability is why the section links to business cycles. The model explains why booms and slumps can feed on themselves.

5. Use in India: the First Five Year Plan

• The First Five Year Plan (1951-56) was built on Harrod-Domar-type logic: raise saving and investment, and growth follows.
• Target: 2.1% a year. Achieved: about 3.6%.
• Why it fit the India of 1951: saving was low and capital was scarce. So "capital shortage" looked like the main barrier to growth.

6. Limits of Harrod-Domar

• Fixed coefficients: capital and labour are used in fixed proportions and cannot be substituted. A labour-rich country like India cannot use more workers in place of scarce machines.
• No technical progress: better technology plays no role inside the model.
• Saving is treated as the only constraint. The model ignores skills, demand and institutions (law, governance, markets).
• It explains little about efficiency. It says how much to invest, not how well. This gap led to the Solow model (1956) in Section 3, which allows substitution and makes technology the long-run driver of growth.

7. The savings-investment gap

• Open-economy identity: S − I = X − M.
• S = domestic saving, I = domestic investment, X = exports, M = imports.

• (X − M) here stands for the current account balance: trade in goods and services plus income and transfers.

• If domestic investment is greater than domestic saving, (X − M) turns negative. That is a current account deficit (CAD) (the country buys more from the world than it sells, and borrows the difference).

• The CAD is financed by foreign capital:
• FDI (foreign direct investment: long-term ownership stakes in firms).
• FPI (foreign portfolio investment: purchases of shares and bonds).

• Loans (external commercial borrowing, aid).

• Savings-investment gap = the shortfall of domestic saving against domestic investment. It is mirrored in the CAD.

• Worked example: saving = 30% of GDP and investment = 32% of GDP → gap = 2% of GDP → CAD ≈ 2% of GDP, filled by foreign savings.
• Link to Harrod-Domar: foreign savings raise the "s" available for investment. With ICOR = 4, a CAD of 2% of GDP adds about 2 ÷ 4 = 0.5 percentage point of growth, but it must be repaid or serviced later.

8. India's data

• Saving and investment levels (scaffold; verify from the Economic Survey and RBI): gross domestic saving and gross fixed capital formation (GFCF) (spending on new machines, buildings and infrastructure, net of sales) both sit around 30% of GDP.
• Investment rate after COVID: the investment-to-GDP ratio rose to about 29.6% in 2021-22, the highest in seven years. The government linked this to its push for public capex and infrastructure [3][4].
• GFCF in rupees: GFCF rose from ₹32.78 lakh crore (2014-15) to ₹54.35 lakh crore (2022-23, Provisional Estimates), at constant 2011-12 prices [2].
• Public capex: Centre's capital expenditure was ₹9.5 lakh crore in FY24, up 28.2% year-on-year [4].
• Market financing of investment: primary capital markets raised ₹10.9 lakh crore in FY24. That is about 29% of the GFCF of private and public corporates in FY23 [4].
• Household net financial saving (households' bank deposits, shares, insurance and so on, minus their borrowings) has fallen. It was about 5% of GDP in 2022-23 per RBI, a multi-decade low (verify latest).
• Households are shifting to physical assets (houses, gold).
• Households are taking on more debt (home, vehicle and personal loans).

• Why this matters: household financial saving is the main pool that banks lend on to firms and the government. If it shrinks, investment must rely more on foreign savings.

• ICOR has hovered around 4-5 in recent years (verify).

• With investment near 30% of GDP, this implies growth of roughly 30 ÷ 5 = 6% to 30 ÷ 4 = 7.5%.

• Current account deficit (latest): CAD was US$23.3 billion (0.6% of GDP) in 2024-25, down from US$26.0 billion (0.7% of GDP) in 2023-24, mainly because of higher net invisibles receipts (services exports and remittances) [5].

• Reading it: India's savings-investment gap is now small. Domestic saving funds almost all domestic investment.

9. Financing side and the risk of a large CAD (Class 11, LPG: An Appraisal)

• Foreign investment (FDI + FII) rose from about US$100 million (1990-91) to US$23 billion (2022-23).
• Forex reserves rose from about US$6 billion (1990-91) to about US$646 billion (2023-24).
• Foreign savings can close the gap, but a large, persistent CAD is risky:
• Foreign money can leave quickly, especially FPI ("hot money").

• Debt must be repaid in dollars, so a falling rupee makes repayment costlier.

• 1991 BoP crisis (balance of payments crisis): it began when forex reserves covered less than two weeks of imports. It led to the LPG (liberalisation, privatisation and globalisation) reforms of 1991.

Prelims Hooks

• Harrod-Domar: g = s / v. With s = 30% and v = 4, g = 7.5%. A higher ICOR means lower growth at the same saving rate.
• ICOR = ΔK/ΔY ≈ investment rate ÷ GDP growth rate. A lower ICOR means more efficient capital. This is a common trap: a "higher ICOR = better" statement is wrong.
• Warranted growth rate (Gw) = the rate at which firms' new capacity is fully used. Natural growth rate (Gn) = labour-force growth + technical progress.
• Knife-edge: steady growth needs Gw = Gn = G. Gaps widen, they do not self-correct.
• The First Five Year Plan (1951-56) used Harrod-Domar logic. Target 2.1%, achieved about 3.6%.
• S − I = X − M: if investment > saving, the country runs a current account deficit.
• Harrod-Domar assumes fixed coefficients and no technical progress. The Solow model (1956) relaxes both.
• India's CAD: 0.6% of GDP (US$23.3 bn) in 2024-25 vs 0.7% (US$26.0 bn) in 2023-24 [5].
• 1991 BoP crisis: forex reserves covered less than two weeks of imports.

Mains Points

• Quality of investment matters, not only quantity. With ICOR at 4-5, reaching 8% growth needs investment of 32-40% of GDP. Faster project completion, less idle capacity and better logistics lower ICOR. That is cheaper than pushing the saving rate higher. Public capex of ₹9.5 lakh crore (FY24) helps only if projects finish on time [4].
• The household saving squeeze is a risk to long-run growth. Net financial saving near a multi-decade low (2022-23, RBI) and rising household debt shrink the pool of funds that banks lend to firms. Growth then depends more on foreign savings and a wider CAD. The 1991 crisis shows why a large, persistent CAD is dangerous.
• Harrod-Domar is useful but incomplete for India. Early Plans rightly saw capital as scarce. But the model ignores labour substitution, skills, technology and institutions. A labour-surplus economy needs growth that creates jobs, which points towards Solow and later human-capital approaches.
• The knife-edge helps explain business cycles and post-COVID policy. When demand falls short of capacity, private firms cut investment and the slump deepens. This is the case for counter-cyclical public capex, which lifted the investment rate to about 29.6% in 2021-22 [3][4].