Solow Growth Model: Capital Accumulation, Convergence and the Golden Rule

An economy as a capital accumulation problem

The Solow-Swan growth model (1956) explains long-run economic growth with a small set of moving parts: output is produced from capital and effective labour through a production function, a fixed share of output is saved and reinvested as new capital, and existing capital wears out at a constant rate.

Y = K^alpha * (A*L)^(1 - alpha)     Cobb-Douglas production function

Y = output          K = capital           L = labour
A = technology (labour-augmenting)        alpha = capital's output share (0 < alpha < 1)

dK/dt = s*Y - delta*K              capital accumulation
s = savings rate     delta = depreciation rate

Diminishing returns and the steady state

The exponent α is less than 1, which builds diminishing returns to capital directly into the model: each additional unit of capital per worker raises output by less than the previous unit did, holding technology and labour fixed. Combined with constant depreciation, this produces a stable steady state: capital per effective worker, k = K/(AL), grows whenever new investment (sY) exceeds what is needed to replace depreciating and diluting capital, and shrinks whenever it falls short, so the economy converges to the single level of k where the two exactly balance — convergence, not perpetual acceleration, is the model's central, sometimes counter-intuitive, prediction.

steady state:  s * f(k*) = (delta + n + g) * k*

k*  = steady-state capital per effective worker
n   = population (labour force) growth rate
g   = technology growth rate
f(k) = k^alpha        (per-worker production function)

Why a poor country can 'catch up' faster

Diminishing returns has a sharp implication for growth rates, not just levels: an economy that starts far below its own steady state has a large gap between investment and the capital it needs, so it grows quickly; an economy already near its steady state has little gap left and grows slowly. This is the model's conditional convergence prediction — poorer economies tend to grow faster toward their own steady state, not necessarily toward the same absolute income level as richer economies, since each economy's steady state depends on its own savings rate, population growth and technology level. It is a widely cited (and contested) explanation for why some developing economies have historically grown faster than mature ones.

Why growth doesn't stop at the steady state

Reaching the steady state does not mean the economy stops growing in absolute terms — it means capital per effective worker stops growing. Total output can still rise forever at the steady state, driven by the two things the model holds as exogenous, unexplained inputs: population growth n (more workers) and technological progress g (each worker becomes more productive with the same capital). In the long run, output per worker grows at exactly the rate of technological progress g — the Solow model's famous conclusion that, absent ongoing technological improvement, per-capita income growth eventually grinds to a halt no matter how high the savings rate is, which is why the model treats technology as the ultimate engine of sustained living-standard growth even though it does not explain where that technology comes from.

The golden rule savings rate

A higher savings rate s raises the steady-state level of capital and output, but savings is a trade-off — every unit of output saved is a unit not consumed today. The golden rule savings rate is the specific s that maximises steady-state consumption per worker (not output), found where the extra output from one more unit of steady-state capital exactly equals what it costs to maintain (the marginal product of capital equals δ + n + g). Save less than this and you are leaving free extra long-run consumption on the table; save more, and you are over-accumulating capital, sacrificing present consumption for capital that yields less extra output than it costs to maintain — a genuine possibility the model calls dynamic inefficiency.

Frequently asked questions

Try it live

Everything above runs in your browser — open Solow Growth Model and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.