Solow Growth Model: Steady State, Golden Rule, Convergence

In 30 seconds

The Solow growth model explains how saving, population growth and technology determine a country’s long-run income per worker.
Capital per worker changes by $ \Delta k = s f(k) – (n+\delta)k $: new investment minus the investment needed to equip new workers and replace worn-out machines.
The economy settles at a steady state where $ \Delta k = 0 $. With $ y = k^{\alpha} $, steady-state capital is $ k^* = \left(\frac{s}{n+\delta}\right)^{\frac{1}{1-\alpha}} $.
A higher saving rate raises the level of income but not its long-run growth rate. Only technological progress gives lasting growth.
Exam favourite: the golden rule capital stock maximises consumption per worker and is reached where $ MPK = n + \delta $.

What is the Solow growth model?

The Solow growth model is the standard starting point for explaining why some countries are rich and others are poor, and why living standards rise over time. Robert Solow set it out in his 1956 paper “A Contribution to the Theory of Economic Growth” in the Quarterly Journal of Economics^[1]. Trevor Swan published a very similar model in the same year^[2], so you will sometimes see it called the Solow–Swan model. Solow received the 1987 Nobel Memorial Prize in Economic Sciences “for his contributions to the theory of economic growth”^[3].

The model asks: if a country saves part of its income and invests it in machines, factories and roads, what happens to output per worker over time? Saving alone cannot make an economy grow forever. Each extra machine adds less output than the one before, so the economy reaches a resting point called the steady state. Lasting growth must come from technology.

The key assumptions

One good, closed economy, no government. Saving equals investment: $ I = sY $.
Constant saving rate $ s $. Households save a fixed fraction of income, between 0 and 1. The model does not explain where $ s $ comes from; it is exogenous (given from outside the model).
Labour grows at rate $ n $. The workforce grows at a constant percentage each year.
Capital depreciates at rate $ \delta $. Each year a fixed share $ \delta $ (delta) of the capital stock wears out.
Constant returns to scale. Doubling both capital and labour doubles output.
Diminishing marginal returns to capital. Holding labour fixed, each extra unit of capital adds less output than the last.
Exogenous technology. In the basic model, technology is fixed. In the extended model it grows at a constant rate $ g $ that the model takes as given.

The production function and capital accumulation

Start with an aggregate production function, $ Y = F(K, L) $, where $ Y $ is output, $ K $ is the capital stock and $ L $ is labour. The usual choice is the Cobb-Douglas form:

\[ Y = K^{\alpha} L^{1-\alpha}, \qquad 0 < \alpha < 1 \]

The exponent $ \alpha $ (alpha) measures how much output responds to capital. Under competitive markets it also equals capital’s share of national income. In most rich countries this share is roughly one third, so textbooks usually set $ \alpha = 1/3 $.

Because of constant returns to scale, you can divide everything by $ L $. Write lower-case letters for per-worker values: $ y = Y/L $ is output per worker and $ k = K/L $ is capital per worker. Then:

\[ y = f(k) = k^{\alpha} \]

This per-worker production function rises with $ k $ at a falling rate. Its slope is the marginal product of capital:

\[ MPK = f'(k) = \alpha k^{\alpha – 1} \]

As $ k $ rises, $ MPK $ falls: a poor economy gets a big boost from each new machine, a rich one a small boost. This fact drives most of the model’s results.

Capital accumulation: the fundamental equation

Two forces change capital per worker each year:

Investment adds capital. Investment per worker is $ i = s y = s f(k) $.
Break-even investment keeps $ k $ constant. Some capital wears out ($ \delta k $) and new workers need equipping ($ n k $). Together this is $ (n+\delta)k $. It is the amount of investment needed just to stop $ k $ from falling.

Putting the two together gives the fundamental equation of the Solow model:

\[ \Delta k = s f(k) – (n+\delta)k \]

If actual investment is greater than break-even investment, $ k $ rises. If it is smaller, $ k $ falls. If the two are equal, $ k $ stays still.

Key termBreak-even investment: the investment per worker, $ (n+\delta)k $, that exactly offsets depreciation and the dilution of capital across a growing workforce.

Common mistakeStudents often write $ \Delta k = s f(k) – \delta k $ and forget population growth. If the workforce grows, the same capital stock is spread over more workers, so $ k $ falls even if no machine breaks. That is why $ n $ sits next to $ \delta $.

The steady state: derivation with Cobb-Douglas

The Solow model steady state is the value of capital per worker, $ k^* $, at which investment exactly equals break-even investment, so $ \Delta k = 0 $. Set the fundamental equation to zero:

\[ s (k^*)^{\alpha} = (n+\delta) k^* \]

Divide both sides by $ (k^*)^{\alpha} $:

\[ s = (n+\delta)(k^*)^{1-\alpha} \]

Rearrange for $ (k^*)^{1-\alpha} $ and raise both sides to the power $ 1/(1-\alpha) $:

\[ k^* = \left( \frac{s}{n+\delta} \right)^{\frac{1}{1-\alpha}} \]

Steady-state output per worker follows by substituting back into $ y = k^{\alpha} $:

\[ y^* = \left( \frac{s}{n+\delta} \right)^{\frac{\alpha}{1-\alpha}} \]

Consumption per worker is what is left after saving: $ c^* = (1-s)y^* $.

So countries with a higher saving rate have more capital and income per worker in the steady state; countries with faster population growth or depreciation have less.

Why the steady state is stable

Below $ k^* $, diminishing returns make $ s f(k) $ exceed $ (n+\delta)k $, so capital builds up. Above $ k^* $, break-even investment is larger and $ k $ shrinks. Either way the economy moves towards $ k^* $: a stable equilibrium.

Worked example: finding the steady state

Worked example

Suppose an economy has the production function $ y = k^{1/3} $. Its saving rate is 24% ($ s = 0.24 $), the workforce grows at 2% a year ($ n = 0.02 $) and capital depreciates at 6% a year ($ \delta = 0.06 $). Find steady-state capital, output and consumption per worker. Then check that investment equals break-even investment.

Step 1Work out the ratio: $ \dfrac{s}{n+\delta} = \dfrac{0.24}{0.02+0.06} = \dfrac{0.24}{0.08} = 3 $

Step 2Find the exponent: $ \dfrac{1}{1-\alpha} = \dfrac{1}{1-1/3} = \dfrac{1}{2/3} = 1.5 $

Step 3Steady-state capital: $ k^* = 3^{1.5} = 3 \times \sqrt{3} \approx 5.196 $

Step 4Steady-state output: $ y^* = (k^*)^{1/3} = (3^{1.5})^{1/3} = 3^{0.5} \approx 1.732 $

Step 5Steady-state consumption: $ c^* = (1-0.24) \times 1.732 = 0.76 \times 1.732 \approx 1.316 $

Step 6Check: investment $ s y^* = 0.24 \times 1.732 \approx 0.416 $; break-even $ (n+\delta)k^* = 0.08 \times 5.196 \approx 0.416 $. They match, so $ \Delta k = 0 $.

In words: this economy settles at about 5.2 units of capital per worker, producing about 1.73 units of output per worker, of which about 1.32 units are consumed. Once there, output per worker stops growing.

The transition path

Now suppose the same economy starts with only $ k_0 = 2 $. Output per worker is $ 2^{1/3} \approx 1.260 $. Investment is $ 0.24 \times 1.260 \approx 0.302 $. Break-even investment is $ 0.08 \times 2 = 0.16 $. So $ \Delta k \approx 0.302 – 0.16 = 0.142 $ and next year $ k_1 \approx 2.142 $. Each year $ \Delta k $ shrinks, because extra capital adds less and less output. Growth is fastest when the economy is furthest below its steady state: the seed of convergence.

Exam tipWith $ \alpha = 1/3 $, the exponents become easy: $ k^* = (s/(n+\delta))^{1.5} $ and $ y^* = (s/(n+\delta))^{0.5} $. Choose numbers where $ s/(n+\delta) $ is a perfect square (such as 4 or 9) and your answers come out as whole numbers. Examiners often do the same.

The Solow diagram: a rise in saving and a rise in population growth

DiagramDraw capital per worker $ k $ on the horizontal axis and output or investment per worker on the vertical axis. Draw three lines. (1) The production function $ f(k) $: a curve from the origin that rises steeply at first and then flattens. (2) The investment curve $ s f(k) $: the same shape, but lower, because it is a fraction $ s $ of output. (3) The break-even line $ (n+\delta)k $: a straight line from the origin with slope $ n+\delta $. The investment curve cuts the break-even line at point A, where $ k = k^* $. To the left of A the investment curve lies above the line, so arrows on the horizontal axis point right. To the right of A the curve lies below the line, so arrows point left. The vertical distance from the $ k $ axis up to $ f(k^*) $ is $ y^* $; the gap between $ f(k^*) $ and $ s f(k^*) $ is consumption $ c^* $.

A rise in the saving rate

If $ s $ rises from $ s_1 $ to $ s_2 $, the investment curve pivots up to $ s_2 f(k) $. At the old $ k^* $, investment now exceeds break-even investment, so $ k $ rises until a new intersection at point B, with $ k_2^* > k_1^* $.

Using our example: if $ s $ rises from 0.24 to 0.32, then $ s/(n+\delta) = 0.32/0.08 = 4 $. New $ k^* = 4^{1.5} = 8 $ and new $ y^* = 4^{0.5} = 2 $. Output per worker rises from about 1.73 to 2.

During the move from A to B, output per worker grows faster. Once at B, growth falls back to zero. A higher saving rate causes a level effect, not a permanent growth effect.

A rise in population growth

If $ n $ rises, the break-even line pivots up and becomes steeper. It cuts the investment curve further left, at point C, so $ k^* $ falls. In our example, if $ n $ rises from 0.02 to 0.04, then $ n+\delta = 0.10 $, $ s/(n+\delta) = 2.4 $, $ k^* = 2.4^{1.5} \approx 3.72 $ and $ y^* = 2.4^{0.5} \approx 1.55 $. Capital is spread more thinly, so each worker produces less.

Common mistakeSaying “higher saving makes the economy grow faster in the long run.” In the Solow model it does not. Total output $ Y $ still grows at $ n $ (plus $ g $ once technology is added) in the steady state, whatever the saving rate. Saving shifts the level of the path, not its slope.

The golden rule level of capital

Should a country save as much as possible? No. At a saving rate of 100%, nobody would consume anything, and consumption is what people enjoy. Edmund Phelps called the saving rate that maximises steady-state consumption per worker the “golden rule”^[4].

In the steady state, consumption per worker is output minus break-even investment:

\[ c^* = f(k^*) – (n+\delta)k^* \]

To maximise $ c^* $, differentiate with respect to $ k^* $ and set the result to zero:

\[ f'(k_{gold}) = n + \delta \quad\Longleftrightarrow\quad MPK = n + \delta \]

On the diagram, the golden rule level of capital is where the slope of the production function equals the slope of the break-even line. At that point the vertical gap between $ f(k) $ and $ (n+\delta)k $, which is consumption, is as large as it can be.

With Cobb-Douglas, $ \alpha k^{\alpha-1} = n+\delta $ gives $ k_{gold} = \left(\frac{\alpha}{n+\delta}\right)^{\frac{1}{1-\alpha}} $. Compare this with the steady-state formula and you can see that the golden rule saving rate is simply $ s_{gold} = \alpha $. When capital’s share of income is one third, the golden rule saving rate is one third.

Worked example

Using $ y = k^{1/3} $, $ n = 0.02 $ and $ \delta = 0.06 $, find the golden rule capital stock and consumption, and compare them with saving rates of 24% and 32%.

Step 1Golden rule saving rate: $ s_{gold} = \alpha = 1/3 \approx 0.333 $

Step 2$ k_{gold} = \left(\dfrac{1/3}{0.08}\right)^{1.5} = 4.167^{1.5} \approx 8.51 $

Step 3$ y_{gold} = 4.167^{0.5} \approx 2.041 $, so $ c_{gold} = (1 – 1/3) \times 2.041 \approx 1.361 $

Step 4Check: $ MPK = \tfrac{1}{3} \times 8.51^{-2/3} = \tfrac{1}{3} \div 4.167 \approx 0.08 = n+\delta $

Step 5Compare: at $ s = 0.24 $, $ c^* \approx 1.316 $; at $ s = 0.32 $, $ c^* = 0.68 \times 2 = 1.360 $. Both are below 1.361.

In words: consumption per worker peaks when one third of income is saved. Saving 24% leaves too little capital. Saving more than one third leaves too much: so much output goes on maintaining capital that consumption falls.

Exam tipIf an economy has more capital than the golden rule level (it is “dynamically inefficient”), cutting saving raises consumption both now and in the future, so everyone gains. If it has less, raising saving means lower consumption today in exchange for higher consumption later. This trade-off between generations is a strong evaluation point.

Adding technology: balanced growth

The basic model predicts that output per worker eventually stops growing, yet in rich countries it has grown for over a century. Solow’s own answer was technological progress. In a 1957 study of the United States, he estimated that output per hour of work doubled between 1909 and 1949, and that about seven-eighths (87.5%) of that rise came from technical change rather than more capital per worker^[5]. The unexplained part became known as the “Solow residual” or total factor productivity.

To add technology, write the production function with labour-augmenting technology $ A $, which grows at rate $ g $:

\[ Y = K^{\alpha}(AL)^{1-\alpha} \]

Here $ AL $ is “effective labour”: the number of workers multiplied by how productive each one is. Now define capital per effective worker, $ \tilde{k} = K/(AL) $. The accumulation equation becomes:

\[ \Delta \tilde{k} = s\tilde{k}^{\alpha} – (n + g + \delta)\tilde{k} \]

The steady state is the same as before with $ g $ added to the break-even rate: $ \tilde{k}^* = \left(\frac{s}{n+g+\delta}\right)^{\frac{1}{1-\alpha}} $.

In this steady state, capital and output per effective worker are constant. Because $ A $ grows at $ g $, output and capital per actual worker grow at $ g $ and total output grows at $ n + g $. The capital–output ratio stays constant. This is balanced growth: key ratios settle while living standards rise at the rate of technological progress.

Key termBalanced growth path: a steady state in which output, capital and consumption per worker all grow at the same constant rate $ g $, the rate of technological progress.

Convergence: theory and real evidence

Diminishing returns imply that an economy far below its steady state grows quickly. There are two versions of this idea.

Absolute (unconditional) convergence: poor countries grow faster than rich ones, full stop. This holds only if all countries share the same steady state.
Conditional convergence: each country converges to its own steady state, set by its saving rate, population growth and technology. A poor country grows fast only if it is poor relative to its own steady state.

What the evidence says

Mankiw, Romer and Weil (1992). In “A Contribution to the Empirics of Economic Growth”, also in the Quarterly Journal of Economics, they asked whether the Solow model fits international data^[6]. They found that saving and population growth affect income in the directions the model predicts. When they added human capital (education and skills) as a second type of capital, the “augmented Solow model” explained about 80% of the international variation in income per capita. They also found that, holding population growth and capital accumulation constant, countries converge at about the rate the augmented model predicts^[6].

Barro (1991). Using data for 98 countries from 1960 to 1985, Robert Barro found that growth in GDP per person was negatively related to the starting level of GDP per person once initial human capital was held constant^[7]. Poor countries with good schooling tended to catch up.

Barro and Sala-i-Martin (1992). Across US states, they found clear convergence at roughly 2% a year^[8]. Across 98 countries from 1960 to 1985, there was no absolute convergence: poorer countries did not, on average, grow faster. Once they held constant differences in schooling, government consumption and other steady-state factors, a convergence rate close to the US-state estimate reappeared^[8]. This is conditional convergence in action.

A 2% convergence rate means an economy closes about 2% of the gap to its steady state each year. The time to close half the gap is roughly $ \ln 2 / 0.02 \approx 35 $ years. Convergence is real, but slow.

A look at recent data

The table and chart below use World Bank data for seven economies. They compare the average investment share of GDP between 2000 and 2023 with the average annual growth of real GDP per person over the same period.

Economy	Average investment, % of GDP (2000–2023)^[9]	GDP per person, 2000 (PPP, constant 2021 international dollars)^[10]	GDP per person, 2023 (PPP, constant 2021 international dollars)^[10]	Average annual growth, 2000–2023
China	41.9	4,066	22,687	7.8%
India	34.1	3,010	8,871	4.8%
South Korea	31.8	26,404	54,029	3.2%
Singapore	25.6	68,904	130,297	2.8%
United States	21.3	55,046	74,352	1.3%
United Kingdom	18.2	43,818	53,423	0.9%
Pakistan	15.9	3,532	5,419	1.9%

Investment is gross capital formation as a share of GDP. Growth is the compound annual rate calculated from the 2000 and 2023 values.

Investment share and growth of GDP per person, 2000–2023 (%)

Average investment, % of GDP (2000–2023)Average annual growth of GDP per person, PPP (2000–2023)

Source: World Bank, World Development Indicators [9][10] Investment is gross capital formation as a share of GDP, averaged over 2000–2023. Growth is the compound annual rate of real GDP per person (PPP, constant 2021 international $) between 2000 and 2023.

The pattern fits the Solow model. China and India started poor and invested heavily, so they were far below high steady states and grew fast. The United States and the United Kingdom were already rich and grew slowly. Pakistan is the useful contrast. It started poorer than China, yet had the lowest investment share, pointing to a lower steady state, and grew more slowly than far richer South Korea. That is what conditional convergence predicts.

Seven countries cannot prove a theory, and investment and growth affect each other both ways, so treat this as an illustration, not a test.

Limitations of the Solow model and endogenous growth

The Solow model is a superb teaching tool, but it has clear weaknesses.

Technology is unexplained. Long-run growth depends entirely on $ g $, which falls from the sky. The model says growth comes from technology but not where technology comes from.
The saving rate is fixed. Households do not choose $ s $ in response to interest rates or policy.
Capital differences alone are too small. With $ \alpha = 1/3 $, realistic differences in saving and population growth predict smaller income gaps than we see. Mankiw, Romer and Weil needed human capital to close much of the gap^[6].
Convergence is too fast in the basic model. With $ \alpha = 1/3 $, the model implies convergence at roughly $ (1-\alpha)(n+g+\delta) $. With $ n+g+\delta $ around 0.08, that is over 5% a year, more than double the 2% found in the data^[8]. A larger capital share, including human capital, slows it down.
Institutions are missing. Property rights, the rule of law and conflict affect investment and technology but sit outside the model.

Endogenous growth in brief

Endogenous growth theory, from the late 1980s, explains technological progress inside the model. Paul Romer (1986) showed that if knowledge spills over between firms, returns to capital need not diminish for the economy as a whole, so growth can continue without outside help^[11]. Robert Lucas (1988) stressed human capital accumulation as an engine of growth^[12]. Romer (1990) then modelled new ideas as the product of profit-seeking research and development^[13]. Here, policies on education and research can change the long-run growth rate, not just the level of income.

Exam tipA strong evaluation paragraph balances both sides: the Solow model correctly predicts that saving and population growth affect income levels and that conditional convergence occurs, but it cannot explain the source of long-run growth. Name endogenous growth theory as the response.

Practice questions

EasyQ1. In the Solow model, what is break-even investment and why does it depend on $ n $? Show worked answer

Break-even investment is $ (n+\delta)k $, the investment per worker needed to keep capital per worker constant. It covers depreciation ($ \delta k $) and the capital needed to equip new workers ($ nk $). If the workforce grows and no extra capital is added, the same capital is spread over more people, so capital per worker falls.

EasyQ2. According to the Solow model, what is the long-run effect of a permanent rise in the saving rate on (a) the level and (b) the growth rate of output per worker, with no technological progress? Show worked answer

(a) The level rises: steady-state $ k^* $ and $ y^* $ are higher. (b) The long-run growth rate is unchanged at zero. Growth is temporarily faster during the transition to the new steady state, then returns to zero.

MediumQ3. An economy has $ y = k^{1/2} $, $ s = 0.2 $, $ n = 0.01 $ and $ \delta = 0.04 $. Find $ k^* $, $ y^* $ and $ c^* $. Show worked answer

$ s/(n+\delta) = 0.2/0.05 = 4 $. With $ \alpha = 1/2 $, $ 1/(1-\alpha) = 2 $, so $ k^* = 4^2 = 16 $. $ y^* = 16^{1/2} = 4 $. $ c^* = (1-0.2) \times 4 = 3.2 $. Check: $ s y^* = 0.8 $ and $ (n+\delta)k^* = 0.05 \times 16 = 0.8 $.

MediumQ4. For the economy in Q3, find the golden rule capital stock and consumption per worker. Is the economy saving too much or too little? Show worked answer

Golden rule: $ MPK = \tfrac{1}{2}k^{-1/2} = 0.05 $, so $ k^{-1/2} = 0.1 $ and $ k_{gold} = 100 $. Then $ y_{gold} = 10 $, and $ s_{gold} = \alpha = 0.5 $, so $ c_{gold} = 0.5 \times 10 = 5 $. Check: $ c = f(k) – (n+\delta)k = 10 – 0.05 \times 100 = 5 $. The economy has $ k^* = 16 < 100 $ and $ c^* = 3.2 < 5 $, so it saves too little relative to the golden rule.

ExamQ5. “Poor countries should grow faster than rich countries.” Evaluate this statement using the Solow growth model and empirical evidence. [15 marks] Show worked answer

Knowledge and understanding (up to 4 marks): state the accumulation equation $ \Delta k = s f(k) – (n+\delta)k $; explain diminishing returns, so $ MPK $ is high when $ k $ is low; define steady state and the difference between absolute and conditional convergence.

Analysis (up to 5 marks): use the Solow diagram to show that an economy far below $ k^* $ has a large gap between investment and break-even investment, so $ k $ and $ y $ grow fast. Show that $ k^* = (s/(n+\delta))^{1/(1-\alpha)} $ differs across countries, so a poor country with low $ s $ or high $ n $ may already be close to a low steady state and grow slowly.

Evaluation (up to 6 marks): evidence shows no absolute convergence across a broad sample of countries (Barro and Sala-i-Martin, 1992) but conditional convergence at about 2% a year once steady-state factors are held constant; Mankiw, Romer and Weil (1992) find the augmented model with human capital explains about 80% of income differences; recent examples such as China and India support fast catch-up with high investment, while low-investment economies converge slowly. Limitations: technology is exogenous, institutions are missing, and convergence in the basic model is faster than observed. Judgement: the statement is true only conditionally, for poor countries that share the saving, population and technology characteristics of rich ones.

Frequently asked questions

What is the main idea of the Solow growth model?

Output per worker depends on capital per worker, but capital faces diminishing returns. Saving and investment therefore raise income only up to a steady state. Long-run growth in living standards comes from technological progress, which the basic model treats as given.

How do you find the steady state in the Solow model?

Set investment equal to break-even investment: $ s f(k) = (n+\delta)k $. With $ y = k^{\alpha} $, solve to get $ k^* = (s/(n+\delta))^{1/(1-\alpha)} $, then $ y^* = (k^*)^{\alpha} $.

What is the golden rule level of capital?

It is the steady-state capital per worker that maximises consumption per worker. It occurs where the marginal product of capital equals $ n+\delta $ (or $ n+g+\delta $ with technology). With Cobb-Douglas production, the golden rule saving rate equals $ \alpha $.

Does the Solow model predict that poor countries catch up?

Only conditionally. Poor countries grow faster if they are far below their own steady state. Countries with low saving or high population growth have low steady states, so they may stay poor. The evidence supports conditional, not absolute, convergence.

References

Solow, R. M. (1956). “A Contribution to the Theory of Economic Growth”. Quarterly Journal of Economics, 70(1), 65–94. Oxford University Press. https://doi.org/10.2307/1884513
Swan, T. W. (1956). “Economic Growth and Capital Accumulation”. Economic Record, 32(2), 334–361. Wiley. https://doi.org/10.1111/j.1475-4932.1956.tb00434.x
Nobel Prize Outreach (1987). “The Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel 1987: Robert M. Solow”. NobelPrize.org. https://www.nobelprize.org/prizes/economic-sciences/1987/summary/
Phelps, E. S. (1961). “The Golden Rule of Accumulation: A Fable for Growthmen”. American Economic Review, 51(4), 638–643. American Economic Association. https://www.jstor.org/stable/1812790
Solow, R. M. (1957). “Technical Change and the Aggregate Production Function”. Review of Economics and Statistics, 39(3), 312–320. MIT Press. https://doi.org/10.2307/1926047
Mankiw, N. G., Romer, D. and Weil, D. N. (1992). “A Contribution to the Empirics of Economic Growth”. Quarterly Journal of Economics, 107(2), 407–437. Working paper version: NBER Working Paper 3541. https://www.nber.org/papers/w3541
Barro, R. J. (1991). “Economic Growth in a Cross Section of Countries”. Quarterly Journal of Economics, 106(2), 407–443. Oxford University Press. https://doi.org/10.2307/2937943
Barro, R. J. and Sala-i-Martin, X. (1992). “Convergence”. Journal of Political Economy, 100(2), 223–251. University of Chicago Press. https://dash.harvard.edu/bitstream/handle/1/3451299/barro_convergence.pdf
World Bank (2026). “Gross capital formation (% of GDP)” (NE.GDI.TOTL.ZS). World Development Indicators, last updated 8 October 2026. https://data.worldbank.org/indicator/NE.GDI.TOTL.ZS
World Bank (2026). “GDP per capita, PPP (constant 2021 international dollars)” (NY.GDP.PCAP.PP.KD). World Development Indicators. https://data.worldbank.org/indicator/NY.GDP.PCAP.PP.KD
Romer, P. M. (1986). “Increasing Returns and Long-Run Growth”. Journal of Political Economy, 94(5), 1002–1037. University of Chicago Press. https://doi.org/10.1086/261420
Lucas, R. E. (1988). “On the Mechanics of Economic Development”. Journal of Monetary Economics, 22(1), 3–42. Elsevier. https://doi.org/10.1016/0304-3932(88)90168-7
Romer, P. M. (1990). “Endogenous Technological Change”. Journal of Political Economy, 98(5, Part 2), S71–S102. University of Chicago Press. https://doi.org/10.1086/261725

The Solow Growth Model Explained: Steady State, Golden Rule and Convergence