📚 The Solow Growth Model Explained | 索洛增长模型解析
The Solow growth model, developed by Robert Solow in 1956, is the cornerstone of modern economic growth theory. It explains how capital accumulation, labour force growth, and technological progress interact to determine the long‑run growth path of an economy. By focusing on the steady state and the role of diminishing returns to capital, the model provides a framework for understanding why some countries grow faster than others and what policies might affect long‑run living standards.
索洛增长模型由 Robert Solow 于 1956 年提出，是现代经济增长理论的基石。它解释了资本积累、劳动力增长和技术进步如何相互作用，以决定经济的长期增长路径。通过聚焦稳态和资本边际报酬递减的作用，该模型为理解为何一些国家比另一些国家增长更快，以及什么样的政策可能影响长期生活水平提供了框架。
1. Origins and Core Assumptions | 起源与核心假设
Robert Solow constructed the model to address shortcomings of the Harrod‑Domar model, which assumed fixed capital‑output ratios. Solow introduced a neoclassical production function with substitution between capital and labour, and he emphasised diminishing marginal returns to each input. The core assumptions are: a closed economy with no government, a single good produced, competitive markets, exogenous saving rate, constant population growth rate n, and constant rate of labour‑augmenting technological progress g. Depreciation occurs at a constant rate δ. These simplifications allow tractable analysis of the economy’s convergence to a balanced growth path.
Robert Solow 构建该模型是为了弥补哈罗德‑多马模型中固定资本‑产出比的缺陷。索洛引入了资本与劳动可互相替代的新古典生产函数，并强调每种要素的边际报酬递减。核心假设包括：封闭经济且无政府部门，只生产一种产品，市场完全竞争，储蓄率外生给定，人口增长率 n 恒定，劳动增强型技术进步率 g 恒定，资本按固定折旧率 δ 消耗。这些简化使得对经济向平衡增长路径收敛的分析易于处理。
2. The Production Function | 生产函数
The aggregate output is given by Y = F(K, AL), where K is capital, L is labour, and A is the level of labour‑augmenting technology. Effective labour is AL. The production function exhibits constant returns to scale, meaning doubling both K and AL doubles output. It also satisfies the Inada conditions: marginal product of capital approaches infinity as capital per effective worker tends to zero, and approaches zero as capital per effective worker becomes very large. In intensive form, we define output per effective worker y = Y/(AL) and capital per effective worker k = K/(AL). Then y = f(k), with f(0) = 0, f'(k) > 0, and f”(k) < 0. A commonly used specification is the Cobb‑Douglas form: Y = Kα (AL)1−α, which in intensive form becomes y = kα, with 0 < α < 1.
总产出由 Y = F(K, AL) 给出，其中 K 是资本，L 是劳动，A 是劳动增强型技术水平。有效劳动为 AL。生产函数具有规模报酬不变的特征，即资本和有效劳动翻倍会使产出翻倍。它还满足稻田条件：当有效劳动人均资本趋于零时，资本边际产出趋于无穷；当有效劳动人均资本很大时，边际产出趋于零。在集约形式中，我们定义有效劳动人均产出 y = Y/(AL) 和有效劳动人均资本 k = K/(AL)。于是 y = f(k)，满足 f(0) = 0，f'(k) > 0 且 f”(k) < 0。常用的设定是柯布‑道格拉斯形式：Y = Kα (AL)1−α，集约形式为 y = kα，0 < α < 1。
3. Capital Accumulation Equation | 资本积累方程
The change in the capital stock over time comes from gross investment minus depreciation. Gross investment equals saving, where the saving rate s is a constant fraction of output. The accumulation of total capital is: ΔK = sY − δK. To express the dynamics in terms of effective‑worker variables, we differentiate k = K/(AL). The result is the fundamental equation of the Solow model:
Δk = s f(k) − (δ + n + g) k
This equation states that the change in capital per effective worker equals actual investment per effective worker sf(k) minus the amount of investment needed to keep k constant due to depreciation, population growth, and technological progress. The term (δ + n + g)k is called break‑even investment. When actual investment exceeds break‑even investment, k rises; when it falls short, k declines.
资本存量随时间的变动等于总投资减去折旧。总投资等于储蓄，储蓄率 s 是产出的一个固定比例。总资本积累方程为：ΔK = sY − δK。为用有效劳动人均变量表示动态变化，我们对 k = K/(AL) 求导，得到索洛模型的基本方程：
Δk = s f(k) − (δ + n + g) k
该方程表明，有效劳动人均资本的变化等于有效劳动人均实际投资 sf(k) 减去因折旧、人口增长和技术进步所需维持 k 不变的投资量。项 (δ + n + g)k 被称为持平投资。当实际投资超过持平投资时，k 上升；反之，k 下降。
4. The Steady State | 稳态
A steady state is defined as a situation where capital per effective worker, output per effective worker, and consumption per effective worker are constant over time. Setting Δk = 0 yields the steady‑state condition:
s f(k*) = (δ + n + g) k*
At k*, the two curves intersect. Because of diminishing returns to capital, the sf(k) curve is concave while the (δ + n + g)k line is straight. Uniqueness and stability of the steady state are guaranteed: if the economy starts below k*, actual investment exceeds break‑even investment and k grows; if it starts above k*, k falls. In the steady state, output per effective worker y* = f(k*) is constant. Total output Y grows at the rate (n + g), and output per capita Y/L grows at the rate g, driven entirely by technological progress. This explains why sustained increases in living standards require continuous technological improvement.
稳态指有效劳动人均资本、有效劳动人均产出和有效劳动人均消费均保持不变的状况。令 Δk = 0，可得稳态条件：
s f(k*) = (δ + n + g) k*
在 k* 处两条曲线相交。由于资本的边际报酬递减，sf(k) 曲线是凹的，而 (δ + n + g)k 线是直线。稳态的唯一性和稳定性得到保证：若经济从低于 k* 的位置起步，实际投资超过持平投资，k 上升；若从高于 k* 起步，k 下降。稳态下，有效劳动人均产出 y* = f(k*) 不变。总产出 Y 以 (n + g) 的速度增长，人均产出 Y/L 以速度 g 增长，完全由技术进步驱动。这就解释了为什么持续的生活水平提高需要持续的技术进步。
5. Impact of Saving Rate on Steady State | 储蓄率对稳态的影响
A once‑and‑for‑all increase in the saving rate shifts the sf(k) curve upwards. In the short run, this raises investment and causes k to rise towards a new, higher steady state. During the transition, output per effective worker grows faster than g, and output per capita grows faster than g. However, as k approaches the new k*, the growth rate gradually falls back to g. Thus, a higher saving rate permanently raises the level of output per effective worker and output per capita, but it does not affect the long‑run growth rate. In the Cobb‑Douglas case y = kα, the steady‑state capital per effective worker is k* = (s / (δ+n+g))1/(1−α). This shows that k* is increasing in s and decreasing in (δ+n+g).
储蓄率一次性提高会使 sf(k) 曲线向上移动。短期内，投资增加，k 向新的更高的稳态增长。在转型过程中，有效劳动人均产出增速高于 g，人均产出增速也高于 g。但当 k 接近新的 k* 时，增长率逐渐回落至 g。因此，较高的储蓄率虽然能永久性地提高有效劳动人均产出和人均产出的水平，但并不影响长期增长率。在柯布‑道格拉斯情形 y = kα 下，稳态有效劳动人均资本为 k* = (s / (δ+n+g))1/(1−α)。这表明 k* 随 s 增加而增加，随 (δ+n+g) 增加而减少。
6. The Golden Rule Level of Capital | 黄金律资本水平
Steady‑state consumption per effective worker is given by c* = f(k*) − s f(k*) = f(k*) − (δ + n + g) k*. While a higher saving rate raises steady‑state output, it may not maximise consumption because a larger share of output must be devoted to investment. The golden rule level of capital kgold maximises steady‑state consumption. It is found by setting the derivative of c* with respect to k* to zero, which yields:
f'(kgold) = δ + n + g
This says the marginal product of capital should equal the break‑even investment rate. If the actual capital stock is below the golden rule, increasing saving can raise long‑run consumption; if it is above, the economy is dynamically inefficient — reducing saving would raise consumption both during the transition and in the new steady state. The golden rule provides a normative benchmark for assessing whether an economy’s saving rate is too high or too low.
稳态有效劳动人均消费由 c* = f(k*) − s f(k*) = f(k*) − (δ + n + g) k* 给出。较高的储蓄率虽然能提高稳态产出，但不一定使消费最大化，因为产出中必须用于投资的份额更大。黄金律资本水平 kgold 使稳态消费最大化。将 c* 对 k* 求导并置零可得：
f'(kgold) = δ + n + g
即资本的边际产出应等于持平投资率。若实际资本存量低于黄金律水平，增加储蓄可以提高长期消费；若高于黄金律水平，经济处于动态无效率状态——减少储蓄既能在转型期也能在新的稳态下提高消费。黄金律为判断一国储蓄率是否过高或过低提供了规范性的基准。
7. Introducing Technological Progress | 引入技术进步
Without technological progress, per capita growth would eventually halt as the economy settles into a steady state with constant capital and output per worker. By incorporating labour‑augmenting technological change at rate g, the model generates sustained growth in per capita income. In the steady state, y = Y/(AL) and k = K/(AL) are constant, so Y/L = A y grows at rate g. The nature of technological progress is exogenous in the basic Solow framework — it arrives as a gift from outside the model. This exogeneity is both a strength, as it focuses attention on capital accumulation, and a limitation, as it does not explain the sources of technological innovation. Later endogenous growth models attempt to endogenise A by linking it to R&D, human capital, and spillovers.
如果没有技术进步，随着经济达到稳态，有效劳动人均资本和产出不再增长，人均产出增长最终也将停滞。通过引入速度为 g 的劳动增强型技术进步，模型能使人均收入持续增长。在稳态下，y = Y/(AL) 和 k = K/(AL) 均不变，因此 Y/L = A y 以速度 g 增长。在基本的索洛框架中，技术进步的性质是外生的——它像礼物一样从模型外部到来。这种外生性既是优点，因为它使分析集中于资本积累，也是局限，因为它没有解释技术创新的来源。后来的内生增长模型试图将 A 内生化，将其与研究开发、人力资本和溢出效应联系起来。
8. Convergence Hypothesis | 收敛假说
The Solow model predicts conditional convergence: countries with similar structural parameters (s, n, δ, g) and similar production functions should converge to the same steady‑state level of capital and output per effective worker. Poorer countries grow faster than richer ones during the transition, provided they have similar steady‑state determinants. Unconditional convergence — that all countries converge to the same income level regardless of differences — is not predicted and is contradicted by data. Empirically, there is strong evidence for conditional convergence among OECD countries and in broader samples when controlling for investment rates and population growth. The speed of convergence is roughly 2% per year, implying it takes about 35 years to close half the gap to the steady state.
索洛模型预测条件收敛：拥有相似结构参数（s, n, δ, g）和相似生产函数的国家，其有效劳动人均资本和产出会收敛到相同的稳态水平。在转型过程中，穷国增长速度快于富国，前提是它们的稳态决定因素相似。无条件收敛——即所有国家不论差异都收敛到同一收入水平——并非模型的预测，也与数据矛盾。经验上，在控制投资率和人口增长后，OECD 国家之间以及更广泛的样本中存在条件收敛的有力证据。收敛速度大约为每年 2%，意味着消除与稳态之间一半的差距需要约 35 年。
9. Extensions: Human Capital | 扩展：人力资本
Mankiw, Romer and Weil (1992) extended the Solow model by adding human capital as a separate factor of production. The augmented production function is Y = Kα Hβ (AL)1−α−β, where H is the stock of human capital. By treating investment in education similarly to physical capital, the model accounts for a significant portion of cross‑country income differences. Including human capital reduces the estimated share of physical capital, bringing it more in line with microeconomic evidence, and it improves the model’s ability to explain variation in income per capita while preserving conditional convergence predictions.
Mankiw、Romer 和 Weil（1992）通过将人力资本作为独立生产要素加入而扩展了索洛模型。扩展后的生产函数为 Y = Kα Hβ (AL)1−α−β，其中 H 是人力资本存量。将教育投资视同物质资本后，模型能够解释跨国人均收入差异中的很大一部分。纳入人力资本降低了物质资本份额的估计值，使之更符合微观经济证据，同时增强了模型解释人均收入差异的能力，并保留了条件收敛的预测。
10. Empirical Evidence and the Solow Residual | 经验证据与索洛残差
When applied to data, the basic Solow model explains only a fraction of observed growth if technology is treated as a residual. Most growth accounting exercises find that the Solow residual — total factor productivity growth — accounts for the bulk of output growth in advanced economies. The residual is computed as the growth rate of output minus the weighted growth rates of capital and labour inputs. The key coefficients on capital and labour are often taken from factor shares. Technological progress, broadly defined, emerges as the dominant driver. Cross‑sectional regressions à la Mankiw‑Romer‑Weil show that investment in physical and human capital and population growth can explain about 80% of the variation in income per capita, giving strong support to the augmented Solow model.
当应用于实际数据时，若将技术视为残差，基本索洛模型只能解释观测增长的一小部分。多数增长核算研究发现，索洛残差——全要素生产率增长——是发达经济体产出增长的主要来源。残差通过产出增长率减去资本和劳动投入的加权增长率计算得到。资本与劳动的关键系数通常取自要素份额。广义的技术进步成为主导驱动力。采用 Mankiw‑Romer‑Weil 方式的截面回归表明，物质资本和人力资本的投资率及人口增长率可以解释人均收入差异的约 80%，有力支持了扩展的索洛模型。
11. Criticisms and Limitations | 批评与局限
Despite its elegance, the Solow model has well‑known limitations. It treats technological progress as exogenous, thus leaving the ultimate engine of growth unexplained. It also assumes identical production functions, saving rates, and population growth parameters across diverse economies, which oversimplifies reality. The model predicts that policies affecting saving or population growth have only level effects, not growth effects; this conflicts with evidence that some policy changes appear to have permanent growth consequences. Furthermore, the neglect of institutional factors, trade, and the role of government limits its applicability to many real‑world questions. Nonetheless, the Solow framework remains the indispensable starting point for growth analysis.
尽管索洛模型极为精致，但其局限也众所周知。它将技术进步视为外生，从而未解释增长的最终引擎。它还假设不同经济体具有相同的生产函数、储蓄率和人口增长参数，这过度简化了现实。模型预测，影响储蓄或人口增长的政策仅具有水平效应而非增长效应；这与一些政策变化似乎具有永久性增长后果的证据相矛盾。此外，忽视制度因素、贸易和政府的作用，也限制了它对许多现实问题的适用性。然而，索洛框架依然是增长分析不可或缺的起点。
12. Policy Implications | 政策启示
Although the Solow model suggests that raising the saving rate cannot permanently raise the growth rate of per capita income, it can sustain a higher level of income. Policies that encourage private saving or public investment in infrastructure can therefore deliver lasting welfare gains. The key policy message is that long‑run improvements in living standards ultimately depend on technological progress. Governments can affect technology through funding for basic research, education, and maintaining a business environment that fosters innovation and diffusion. The golden rule also cautions against excessive capital accumulation that might crowd out consumption. Finally, convergence results imply that poorer countries can catch up if they emulate the structural characteristics of richer economies, but institutional and cultural differences remain crucial.
尽管索洛模型表明，提高储蓄率无法永久性地提升人均收入增长率，但却能支撑更高的收入水平。因此，鼓励私人储蓄或对基础设施进行公共投资的政策可以带来持久的福利改进。模型最重要的政策信息是，生活水平的长期改善最终取决于技术进步。政府可以通过资助基础研究、教育以及营造有利于创新与扩散的商业环境来影响技术。黄金律也提醒要避免过度资本积累挤占消费。最后，收敛结果意味着，若穷国拥有与富国相似的结构性特征，它们可以迎头赶上；但制度和文化上的差异仍然至关重要。
Published by TutorHao | Economics Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
屏轩国际教育cambridge primary/secondary checkpoint, cat4, ukiset,ukcat,igcse,alevel,PAT,STEP,MAT, ibdp,ap,ssat,sat,sat2课程辅导，国外大学本科硕士研究生博士课程论文辅导