The basic model predicts that in the long run, growth in output per person stops. This doesn’t match reality—many countries have experienced sustained growth for over a century. To explain this, the Solow model must be expanded to include population growth and, most importantly, technological progress.

Introducing More Moving Parts

When population grows at a rate n and technology improves at a rate g, the steady-state condition becomes more demanding. Investment must now be sufficient to cover depreciation, provide capital for new workers, and keep up with advancing technology. This is often called the “break-even” level of investment.

Think of the capital stock per effective worker, k, as water in a bucket. Every year, a fraction δ leaks out due to depreciation. The bucket also gets wider because there are n new workers who need their own share of water, and it gets taller because technology g makes each worker more productive, requiring more water to keep the level constant. Therefore, to simply keep the water level k from falling—to break even—investment must be sufficient to plug the depreciation leak and fill the new space created by population and technology growth. That break-even amount is (n + g + δ)k.

The Power of Technology

The model incorporates “labor-augmenting” (or “Harrod-neutral”) technological progress. This is a specific type of progress that increases output in the same way as if the economy had more labor. Essentially, technology makes each worker more effective.

Why focus only on ‘labor-augmenting’ technological progress? It might seem arbitrary, but it’s a deliberate choice backed by a crucial mathematical insight from Uzawa’s Theorem. This theorem proves that for an economy to have the “balanced growth” we see in the long run (where key ratios like capital-to-output are stable), technological progress must take this labor-augmenting form. This isn’t just a simplifying assumption; it’s a necessary condition for the model to align with long-run empirical facts.

With technological progress in the picture, we introduce the concept of the effective capital-labor ratio, k = K/(AL). This term is abstract, so let’s use an analogy. Think of A (technology) as a skill multiplier for each worker. If A doubles, it’s as if each worker became twice as effective. The variable k = K/AL doesn’t measure capital per person, but capital per effective unit of labor.

The model’s key prediction is that while output per person (Y/L) can grow forever, the amount of capital per effective worker (k) must eventually stabilize on a balanced growth path. This is how the model generates sustained growth in living standards while still converging to a steady state in its transformed variables. In the long run, only technological progress (g) can explain continuously rising standards of living.

Synthesizing the Full Model

In the full Solow model, investment per effective worker must be sufficient to replenish the effective capital-labor ratio for three key reasons:

1. Depreciation (δ): To replace worn-out capital from the existing stock.
2. Population Growth (n): To equip new workers with the same amount of capital as existing workers.
3. Technological Progress (g): To provide capital for the now “more effective” workforce, keeping the capital-per-effective-worker ratio constant.

With this complete framework—depreciation, population growth, and technology—the Solow model transforms from a theoretical exercise into a diagnostic tool. We can now use its logic to dissect the reasons for the vast income disparities we see across the globe.