To understand the complex dynamics of economic growth, formal theory is indispensable. Theoretical paradigms are essential for organizing facts, clarifying causal relationships, and moving beyond simple accounting exercises to identify the true drivers of growth. An argument about economic prosperity that is not disciplined by a clear theoretical framework is rarely enlightening. This section provides a critical overview of the four major growth paradigms, tracing an intellectual progression that enhances our ability to explain real-world growth patterns.

The Neoclassical Growth Model

The starting point for any study of economic growth is the Neoclassical model developed by Robert Solow and Trevor Swan. Its core mechanism is driven by capital accumulation. However, this accumulation is subject to diminishing marginal returns: as an economy accumulates more capital per worker, each additional unit of capital generates a progressively smaller increase in output.

This core assumption leads to the model’s central prediction: conditional convergence. The theory posits that countries with similar fundamentals—such as saving rates, population growth rates, and depreciation rates—will converge to the same steady-state level of per capita income. A country that starts further below its steady state will grow faster, as it benefits from higher returns to capital, but this rapid growth is purely transitional.

The primary limitation of the Neoclassical model is its inability to explain sustained, long-run growth in per capita income without resorting to an external, unexplained force. To account for the persistent growth observed in advanced economies, the model must assume an exogenous rate of “technical progress” that continually offsets the effect of diminishing returns. In this framework, economic policy can affect the level of income but not the long-run growth rate, which is determined by forces outside the model.

The AK Model’s Contribution to Endogenous Growth

The first wave of endogenous growth theory, the “AK” model, sought to overcome the Neoclassical model’s key limitation. It achieved this by positing that the aggregate production function is linear in a broad measure of capital (Y = AK), effectively assuming constant returns to capital. This broad measure of capital is understood to encompass not just physical capital but also human and intellectual capital. By eliminating diminishing returns, the AK model makes the long-run growth rate endogenous; it now depends on economic decisions, most notably the saving rate. A higher saving rate translates directly into a permanently higher growth rate.

While the AK model successfully endogenized long-run growth, its core assumption of constant returns resolved one limitation at the expense of predictive accuracy, as it failed to account for observed convergence. Its assumption of constant returns implies that growth rates are independent of the initial level of capital, leading to a prediction of divergence or, at best, parallel growth paths for countries with different fundamentals. This is a prediction starkly at odds with the strong empirical evidence of conditional convergence within groups of similar economies, such as across U.S. states (Barro and Sala-i-Martin, 1995b) or OECD nations. Later refinements, such as the open-economy variant by Acemoglu and Ventura (2002), attempted to reintroduce a convergence mechanism through international trade and terms-of-trade effects. This shortcoming motivated the second wave of endogenous growth theory, which shifted the focus from broad capital accumulation to the microeconomics of innovation.

Innovation-Based Growth: Product-Variety vs. Schumpeterian Paradigms

The second wave of endogenous growth theory moved beyond capital accumulation to place innovation at the center of the growth process. Two distinct paradigms emerged from this wave.

The Product-Variety Model

Developed by Paul Romer, this model posits that growth is driven by the creation of new varieties of products. Innovation expands the range of intermediate inputs available for production, leading to increased specialization and higher overall productivity. The incentive for entrepreneurs to innovate is the prospect of earning monopoly rents on their new product. Because aggregate productivity in this model is a function of the number of available varieties (N_t), the exit of a firm reduces N_t and is therefore, by construction, detrimental to growth. The framework lacks a mechanism for “creative destruction,” where obsolescence is a necessary feature of progress.

The Schumpeterian Model

The second branch of innovation-based theory, often called the Schumpeterian model, focuses on quality-improving innovations that render old products obsolete. This is the process Joseph Schumpeter famously termed “creative destruction.” Growth is driven not by adding more products, but by creating better ones. In this model, an innovator who develops a higher-quality intermediate good drives the previous monopolist out of the market and captures its rents until it is, in turn, displaced by a subsequent innovator.

This framework’s most powerful feature, and the reason for its centrality to this monograph, is its natural incorporation of Gerschenkron’s “advantage of backwardness.” For a country operating behind the technological frontier, an “innovation” often consists of successfully implementing a technology that is already in use in more advanced economies. The further a country is from the frontier, the larger the potential quality improvement from such an adoption. Consequently, the potential growth rate depends explicitly on a country’s distance to the frontier.

The Schumpeterian framework, with its focus on creative destruction and the distance-to-frontier effect, provides the most suitable lens for analyzing the nuanced dynamics of convergence, divergence, and the institutional determinants of growth that will be explored next.