In the previous 50 years or so, the gap involving typical living requirements in the poorest nations of the globe and these in the richest nations has grown markedly. This unwelcome modify is contrary to the proposition that significantly less sophisticated economies ought to be in a position to catch up to a lot more sophisticated economies via capital accumulation and technologies transfer. Though the proposition wasn’t new, Abramowitz’s (1986) and Baumol’s (1986) examinations of its veracity as a unfavorable correlation involving initial per capita earnings levels and subsequent development in per capita earnings, initiated a vast literature testing the so-known as convergence hypothesis.1 Following Barro and Sala-i-Martin (1990), the catch-up strategy to the convergence hypothesis became recognized as β-convergence.two 

Table 1 gives some insight into what has driven the widening gap involving the poorest nations and the richest nations. It shows decadal typical per capita GDP development prices for low, upper-middle, and higher-earnings nations (LICs, MICs, and HICs respectively) more than the 50-year period from 1960 to 2010. 

Table 1 Decadal typical per capita GDP development (%) by earnings

Notes: This table is an abridged version of Table two in Johnson and Papageorgiou (2018) exactly where information sources and other facts are provided. There are 29 HICs, 68 MICs, and 51 LICs. Appendix Table A.1 in Johnson and Papageorgiou (2018) lists the nations in every single group.

In spite of the slowdown in their typical development prices more than this period, the HICs have generally grown a lot more swiftly than the MICs, which in turn have grown a lot more swiftly than the LICs. These variations are the precise opposite of the pattern needed for catching up to be observed, and so it is hardly a surprise that the dispersion of per capita GDP across nations has grown markedly considering the fact that 1960. This development is documented in Figure 1, which plots the common deviation of the cross-nation distribution of GDP per capita for a continuous group of nations from 1960 to 2010 and shows that, apart from a smaller decline in the incredibly late 2000s, the dispersion has risen steadily considering the fact that 1960.

In addition to lagging that of the HICs and MICs, the development encounter of the poorest nations from 1960 to 2010 has been heterogeneous, each across nations and more than time. Table 1 shows that the LICs knowledgeable a continuous decline in development prices in each decade from the 1960s to the 1990s, with unfavorable development prices in the 1980s and 1990s, just before the surprising and unprecedented resurgence of development in the 2000s. The MICs saw a comparable resurgence in the 2000s though their slowing in the 1980s did not generate the unfavorable development prices knowledgeable in the LICs.

Table 1 also tends to make the distinction involving what are usually known as fragile and non-fragile LICs. Fragile states are these facing political frailty, characterised by weak institutional capacity, poor governance, corruption, and conflict. This distinction illuminates the marked variations in the development experiences of the fragile LICs and the non-fragile of LICs, most notably in the 1990s and 2000s when the annual typical development prices differ by more than two percentage points. That is, although there is a lot of optimism more than the most current development acceleration in LICs, aggregating their encounter masks the truth that only about half of the LICs are contributing to the resurgence although the rest are stagnant.

Coupled with the steady slowing of development in the HICs, the resurgence in development in the MICs and the non-fragile LICs has been sturdy sufficient to generate the slight decline in the dispersion of per capita GDP across nations starting in the incredibly late 2000s, evident in Figure 1. Responding to Johnson and Papageorgiou (2018), Patel et al. (2018) show that these events are enough to yield a smaller but statistically substantial unfavorable correlation involving development prices and initial earnings more than, for instance, the 1995 to 2010 period. This obtaining of weak β-convergence more than a incredibly brief time period that includes crucial higher and medium-frequency influences on GDP (e.g.  the dot com bubble and its collapse and then the biggest international financial slowdown considering the fact that the 1930s and its quick aftermath) leads them to claim that convergence is occurring. 

As has been recognized for some time, the β-convergence testing strategy is flawed and so the unfavorable correlation observed more than the 1995 to 2010 period does not imply that convergence is basically occurring. In statistical parlance, β-convergence tests can have low energy against non-convergent options. This signifies that such a test can yield a unfavorable correlation even if convergence, which is properly understood as the absence of a part for initial circumstances in figuring out countries’ lengthy-run per capita earnings levels, is not occurring. One particular way to feel about this flaw in the β-convergence strategy is that initial earnings does not completely describe a country’s initial circumstances as there are other crucial aspects that influence development going forward.

Figure 1 Cross-nation earnings dispersion

Supply: Figure six in Johnson and Papageorgiou (2018) exactly where information sources and other facts are provided. 
Notes: Convergence expressed as the decline in the dispersion of per capita GDP across nations is recognized as σ-convergence. As Barro and Sala-i-Martin (1990) show, β-convergence is required but not enough for σ-convergence so the absence of β-convergence for the 1960-2010 period displayed in Table 1 implies the absence of σ-convergence evident in Figure 1. 

An economically fascinating non-convergent option which admits the possibility that initial circumstances can influence lengthy-run per capita earnings levels is that of club convergence. Club convergence happens when the development approach has a lot more than a single steady steady state, as can occur when there is a feedback loop from earnings to some other state variable. In such circumstances, economies converge to the steady state connected with the basin of attraction in which they commence. A country’s lengthy-run per capita earnings will then rely on its initial circumstances as measured by the state variables figuring out its basin of attraction. Nations possessing comparable initial circumstances will converge locally and kind a convergence club but international variations in per capita incomes will persist indefinitely. If there is just a single, steady steady state, initial circumstances will be irrelevant for lengthy-run per capita earnings levels and nations will exhibit catch-up or β-convergence as they gravitate towards it.

Several researchers, like us, have applied a wide variety of empirical approaches in attempts to systematically determine convergence clubs. These approaches involve the use of clustering algorithms in each the cross-section and time series contexts as effectively as projections of the cross-nation distribution of per capita earnings primarily based on its previous evolution. Our reading of this literature (Johnson and Papageorgiou 2018) is that, as soon as the option hypothesis of club convergence is explicitly regarded as, there is sturdy proof contrary to the view that initial circumstances do not matter for lengthy-run outcomes. Alternatively, the information favour the view that initial circumstances do matter and that the cross-nation distribution of per capita earnings exhibits qualities constant with the existence of two or a lot more convergence clubs.three For this purpose, the current optimism in favour of speedy and sustainable convergence appears unfounded. In addition, the development in the dispersion of per capita GDP across nations considering the fact that 1960 documented in Figure 1 suggests that the gaps involving the imply levels of per capita GDP across the convergence clubs are escalating. 

This tendency to neighborhood but not international convergence is illustrated in Figure two which plots the development price more than 1960–2010 against the 1960 level of actual per capita GDP for all of the HICs, MICs, and LICs, with counties in every single group getting shown in distinct colours. The unfavorable correlation involving initial earnings per capita and its subsequent development indicative of catching up is evident inside every single of these 3 groups of nations even although it is not noticed for the nations taken as a single group.

Figure two Development against initial earnings


Supply: Figure four in Johnson and Papageorgiou (2018), exactly where information sources and other facts are provided.

The distinction involving a single single steady steady state and several steady steady states has crucial policy implications due to the fact, in the former case, smaller-scale policy interventions can be valuable to the extent that they hasten the transition of the poor nations to inevitable prosperity. Nevertheless, if there are several steady steady states, then big-scale policy interventions may possibly be needed to push poor economies from a single basin of attraction to an additional. In such a globe, smaller scale policy interventions that fail to push nations into a distinct basin of attraction will at some point be undone as economies fall back to the steady state connected with the basin of attraction in which they lie. A multiplicity of steady states can lead to the sorts of poverty traps discussed by Collier (2007) or the hypothesised middle-earnings trap studied by Eichengreen et al. (2014). The fragile LICs are most likely in such a poverty trap and so need to have a big-scale policy intervention (i.e. a modify that removes their fragility) if they are to move to a steady state with greater per capita earnings. 

The current wave of development in lots of LICs and emerging markets has led to lots of analysts to claim, prematurely in our view, results with slogans such as “lions on the move’ (McKinsey International Institute 2010) or like Patel et al. (2018), to assert that “there has been unconditional convergence.” Nevertheless, with the exception of a handful of nations in Asia which exhibited transformational development, most of the financial achievements in building economies have been the outcome of removing inefficiencies which are merely a single-off level effects. Even though these effects are not unimportant and are required in the approach of improvement, they do not imply ongoing financial development. In addition, we worry that the obstacles that Patel et al. (2018) cite in their conclusion, as effectively as the political fragility of lots of LICs, may possibly undo what progress has lately occurred and bring about lots of LICs to languish in a low-per-capita-earnings steady state.


Abramovitz, M (1986), “Catching Up, Forging Ahead, and Falling Behind,” Journal of Financial History 46(two): 385-406.

Barro, R J and X Sala-i-Martin, (1990), “Financial Development and Convergence Across the United States.” NBER Operating Paper 3419

Baumol, W J (1986), “Productivity Development, Convergence, and Welfare, What the Extended-Run Information Show,” American Financial Review 76(five): 1072-1085.

Collier, P, (2007), The Bottom Billion: Why the Poorest Nations Are Failing and What Can Be Completed About It. Oxford UK: Oxford University Press.

Dowrick, S and J B DeLong. (2003), “Globalization and Convergence”, in M D Bordo, A M Taylor and J G Williamson (eds), Globalization in Historical Viewpoint, NBER Conference Report series, University of Chicago Press.

Durlauf, S N and P A Johnson (1995), “Multiple Regimes and Cross-Nation Development Behavior,” Journal of Applied Econometrics 10(four): 365-284.

Durlauf, S N, P A Johnson, and J R W Temple. (2005), “Growth Econometrics,” in P Aghion and S N Durlauf (Eds.), Handbook of Financial Development, North-Holland, 555-677. 

Eichengreen, B, D Park and K Shin (2014), “Growth slowdowns redux”. Japan and the Globe Economy 32: 65-84. 

Galor, O (2011), Unified Development Theory, Princeton University Press.

Johnson, P, and C Papageorgiou (2018), “What Remains of Cross-Nation Convergence?”, Journal of Financial Literature, forthcoming.

McKinsey International Institute, (2010), Lions On the Move: The Progress and Possible of African Economies. McKinsey &amp Business.

Patel, D, J Sandefur and A Subramanian (2018), “Everything you know about cross-nation convergence is (now) wrong”, PIIE Realtime Financial Challenges Watch, 15 October. 


[1] For surveys of this literature see, amongst other folks, Durlauf et al. (2005), and Johnson and Papageorgiou (2018).

[2] There are two variants of β-convergence: the absolute (or unconditional) variant that considers the correlation involving initial per capita earnings levels and subsequent development in per capita earnings and the conditional variant which considers the correlation right after conditioning on variables believed to decide the steady state worth of output per capita such as prices of physical and human capital accumulation and population development. We are in agreement with Dowrick and DeLong’s (2003) criticism of the latter variant as, in impact, begging the convergence query and so look at only absolute β-convergence.

 [3] Importantly, as Durlauf and Johnson (1995) point out, with the finite spans of information that are obtainable, it is not possible to unequivocally distinguish involving convergence clubs and nations transitioning via distinct stages of improvement just before reaching a frequent steady state as implied by t