General Tariff influencing economic recovery during the 1930s

Evaluate the historical evidence on two mechanisms through which the General Tariff may have influenced the economic recovery during the 1930s.

The General Tariff of 1932 implemented in interwar Britain might be argued to be a significant policy that allowed Britain to recovery strongly after the 1930 recession. Evidence shows that Britain’s recovery was milder and recovery stronger than USA and Germany. There are several mechanisms that the General Tariff could work through to influence the economic recovery including import substitution, effective protection rates (Capie), macroeconomic effects, and the trade blocs. This essay will focus on import substitution and macroeconomic effects, and evidence seems to suggest that tariffs do indeed aid economic recovery in the 1930s.

The import substitution argument suggests that when the general tariff was implemented, imports become more expensive, so people would switch from consuming imports to consuming domestic goods. This results in an increase in net exports (NX) and an increase in domestic consumption (C) that would together lead to an increase in GDP that characterises any economic recovery. A useful framework for evaluating this is the import replacement rate (IRR), where IRR = [del Y – del X]/ del M, so that would be the difference between the change in production and exports divided by the change in imports. If import substitution were effective, then IRR would be close to one, as every unit of production that is not exported offsets a fall in imports directly.

Richardson (1967) used the IRR framework to calculate the effect of tariffs in various industries. He divided them into two groups – the newly protected industries, and the non-newly protected industries. Interestingly, Richardson found that IRR was about 3 for the newly protected industries and IRR = 2 for the non-newly protected industries. This implies that the effect of the tariff was rather weak in import substitution. He uses this premise to later argue that it was the housing boom that led to the recovery.

Vartis and Solomou (1990) criticised Richardson’s methodology. Firstly, even with Richardson’s figures on British trade, exports of newly protected industries increased more and imports fell more than the non-newly protected group. Thus, they argue that Richardson’s measure effectively measures the life cycle of the industry, rather than import substitution. As such, they compared two periods – one before 1930 and another after to observe the effects of the tariff. Then, they found that higher tariffs did lead to higher production, suggesting some grounds for the import substitution argument.

Broadberry and Crafts (2011) conducted a more detailed analysis of the import substitution argument. Instead of the two-way classification that Vartis and Solomou used, Broadberry and Crafts used a three-way classification – they kept the non-newly protected group, and divided the newly protected group into two – those below 10% tariffs and those above. Using a difference in difference methodology, they found that only the newly protected group with tariffs above 10% experienced significant growth, so it was not tariffs itself that mattered, but relative tariffs. Lloyd and Solomou reviewed this article and noted two issues in Broadberry and Crafts’ thesis: (1) there was a methodology issue when we lump the entire 0-10% tariff group together as having no tariffs and 9% tariffs clearly gives different effects (2) there was a classification issue: in particular, rope, twine and bacon clearly had more than 10% tariffs, but Broadberry and Crafts put them in the 0-10% group. With this in mind, Lloyd and Solomou conducted a similar study, and concluded that growth was still positive for the >10% group. Unlikely Broadberry and Crafts who argued that there was no long run productivity differences with additional protection, Solomou and Lloyd suggested that there could be some productivity effects.

Evaluating the import substitution mechanism, evidence seems to suggest that there was positive growth effects on particular industries that had additional protection, though the growth for the economy as a whole through the tariffs might be rather dubious. Hence, we turn to macroeconomic effects next.

Before looking at macroeconomic evidence, it would be useful to understand the theory behind how tariffs can lead to macroeconomic growth. Y = C + I + G + NX, and tariffs have a role in increasing NX and increasing C, due to expenditure switching from exports to imports. For this argument to work, we need evidence for the following: (1) there was an increase in NX and C (2) multiplier effect was positive.

Evidence shows that NX was not a primary cause for Britain’s recovery in the 1930s, because there was effectively a collapse in world trade in the 1930s when almost all countries were protecting. Eichengreen and Irwin (1992), however, qualify that Britain’s trade agreements still put it in a position such that it created trade without diverting trade by having a trade bloc with its empire. Theoretically, if we look at the policy trilemma, we do have good reasons to believe that NX could not increase. The policy trilemma suggests that countries can only have two out of the three objectives: (1) fixed exchange rate (2) independent monetary policy (3) free capital flows. Britain, as the world’s banker, would not restrict capital flows, so it had a choice between (1) and (2). After the devaluation (exit from gold) in 1931, it changed its trilemma position to have an independent monetary policy and forgoing fixed exchange rate. In this position, output Y is determined in the money market, and by comparative statics we know that NX cannot change. If imports were to fall with tariffs, exports have to fall to balance the trade, so exchange rate has to appreciate. The appreciation of exchange rate will work right against the devaluation in 1931. However, we should qualify that Britain was in a managed float system in the 1930s, so it was unlikely to have an uncontrollable downward or upward spiral of exchange rate. This should also be placed in the context of collapsing world trade, so the increase in NX is not a good account.

Consumption increasing is consistent with evidence. In light of low incomes in primary producing countries and in US, much of UK’s recovery in the 1930s was domestically driven, and this would be attributed to the increase in consumption. This is especially true in the housing market: while the housing market took up about 3% of the GDP, it accounted for 17% of the change in GDP between 1932 and 1934, suggesting that consumption here was significant. Richardson noted that a new development block developed here due to the demand created by housing, so this was a strong impetus for recovery. In contrast, Eichengreen notes that less than 10% of the new jobs created during recovery were attributed to any export-linked sector. It should be qualified that while consumption increased, its link to tariffs is rather weak, since most scholars attributed it more to the cheap money policy in 1932.

Keynesians would argue that the multiplier effect during the interwar period was strong. Kahn estimated it to be 1.87 and Keynes between 2 and 3.55. This meant that an exogenous increase in aggregate demand through tariffs can lead to a more than proportionate increase in GDP, and tariffs would then be an effective channel to help with the 1930s recovery.

However, newer evidence suggests otherwise. Crafts and Mills (2012) re-estimated the multiplier during the interwar period and suggested that the multiplier would only be 0.5 to 0.8, due to “psychological crowding out”. Ricardian equivalence occurs when an increase in G results in people expecting taxes to increase in the future, so they would save instead of consume. Since the multiplier effect was contingent on rounds of consumption and high marginal propensity to consume, they would greatly weaken the multiplier. In the context of the interwar period, the government had already incurred large debts from WWI, so any public expenditure would lead to pessimism and more savings rather than spending. As such, it is possible that there is a positive but small multiplier effect, which clearly challenges the role of tariffs.

It is difficult to evaluate the role that tariffs played in Britain’s 1930 recovery as a whole, because the 1932 tariff occurred together with several other policies, and it is difficult to delineate their effects. On the field of import substitution, the effects are clearer because comparative statics can be done across industries as the scholars have done. On the macro-economy, the effects are less clear. Evidence suggests that NX growth was limited, and the recovery was stimulated largely internally, premised on a large housing boom, which can be attributed more to interest rates than tariffs. Estimates on the multiplier are also rather debatable due to the differences in methodologies used. If we look beyond these two mechanisms and into Capie’s study of effective protection rates, and subsequent works by Kitson, Solomou and Weale, we will find that even the effects on individual industries are less clear, and zero correlation was also found by the latter group. As such, historical evidence is not particularly clear on the tariff mechanisms, but it still remains as a useful explanation to understand the 1930s recovery.

Posted in Economics | Leave a comment

Did the Industrial Revolution generate a rapid increase in standards of living for the working class?

The British Industrial Revolution is characterized by the utility of steam, newer technologies, cotton factories etc. Debates on the standard of living (SOL) has been rather polarising, ranging from the optimists who look at the new products and technologies, to the pessimists who look at the unsanitary conditions of urban living that lowered life expectancy. There have been many proxies for the SOL for the working class, including real wages, heights, working hours, life expectancy, and diets. We will look specifically at real wages, heights, and diets in this essay, and it appears that while SOL increased over the industrial revolution, we cannot say that it was indeed “rapid”, so a weak pessimist view should be adopted in light of evidence.

Real wages (W/P) is a very convenient measure of SOL as it gives us a quantitative measure of the people’s purchasing power. An early estimate of this was done by Lindert and Williamson (1987) who looked at workers in different sectors (including agricultural workers, blue collars, and remaining workers) and estimated the real wages of the male workers. They found that W/P was variable for most of the Industrial Revolution (IR) and had a significant increase after 1820. This stagnation in W/P prior to 1820 has been theorised as the Engels Pause. This study then strongly supports the view that there was a “rapid increase” in SOL, if we can take W/P as a good proxy for SOL.

The criticisms immediately following Lindert and Williamson concern the scope of their study. Horrell and Humprhies (1992) argue that they only looked at male wages, which would overstate wages, since males tended to earn more during the IR. Thus, they estimated family incomes instead, and considered people from more sectors, including household weaving. Consequently, they found the trend to be similar to Lindert and Williamson’s, although the extent of W/P increase was less than their estimates. This, however, was highly dependent on Lindert and Williamson’s Cost of living index.

Feinstein (1998) then did the most extensive study to date by considering workers from all sectors, and attempted to revise the cost of living index to gain a better picture of the W/P trend. By looking at the aggregate consumption basket, Feinstein found that real wages were variable up to 1815 and had a slow increase thereafter, so this supports a weak pessimist view.

There were subsequent developments to the real wage data. Clark (2001) argued that Feinstein was too pessimistic, so he re-estimated the consumption basket and the cost of living index, suggesting that the consumption basket for agricultural workers was slightly different. Clark then found that price levels were lower than what Feinstein estimated. Allen (2007) revisited the problem to reconcile the difference between Clark and Feinstein by taking the best measures of each approach, and had estimates that were closer to Feinstein. If we use the progressive credo, evidence seems to suggest a weakly pessimistic view towards W/P. Among the sources, there is agreement that there was some form of Engels pause before 1820, and an increase in W/P thereafter, and the extent of the latter increase is disputed.

Heights appears to be a promising measure of SOL, as it is a composite measure. Food gives people energy, which can be used in the following ways (1) basal metabolism (2) work (3) disease fighting (4) growth. Thus, higher people (as a result of more growth) correlate with less work, less diseases, and more food, which are reasonable proxies for a higher SOL.

Floud et al attempted to measure height trends in IR by looking at heights of military recruits, since such heights data are readily available in their records at enlistment. Looking at the heights of recruits in the Royal Navy and the Marines, Floud et al found that the heights of people born up to 1770 were increasing, and started falling thereafter.

In light of the truncated sample, Nicholas and Oxley did a separate study and found a different result. They looked at the heights of female convicts transported to Australia and observed several trends. Firstly, the heights of people from urban areas were lower than those from rural areas. Secondly, the heights of both groups fell over time, with the rural heights falling more (which is consistent with enclosures that led to the decreasing use of commons and food available to the people without land). Lower heights of people from urban areas is consistent with Szreter and Mooney’s study that life expectancy was lower in urban areas and Huck (1995)’s study that infant mortality was higher, as they all point to the idea that there were more diseases in urban areas.

The use of heights is not without criticism. Correlation with other measures is one major problem. The real wage measures tell us that SOL seemed to be stagnant until 1820 and started increasing thereafter. However, Floud et al’s heights data tell us that SOL increased up to 1820 and fell thereafter. Nicholas and Oxley’s heights data tell us that heights were falling throughout the IR period. Thus, there is a lack of consistency between different methods of measuring heights and with other SOL measures.

Another problem in measuring heights is the time as a proxy for SOL – do people’s current heights tell us about their current SOL, or the SOL at the time of birth? The Barker hypothesis suggests that people’s life expectancy at 50 depends on their month of birth: life expectancy is lower for those whose mothers were pregnant over winter and there was less food available. Cinnirella (2008) looked at this aspect in IR, and compared the heights of people born in rural areas who migrated to cities against those who stayed in rural areas. It was found that the urban migrants were shorter on average – this suggests that heights not only proxy for SOL at the time of birth, but also one’s current living conditions. Cinnirella’s study also shows a steady decrease in heights over time.

The heights data complicate our inferences on SOL from real wage data due to its inconsistency. Nonetheless, it seems that heights were decreasing in the latter half of IR, and a weakly pessimist position is implied.

Another consideration for SOL would be diets and consumption. Muldrew (2011) looked at agricultural produce to estimate the amount of calories available. He found that there was 4000-7500 kcal available per person (and we only need 2500 for our sedentary lifestyle) during the period, which suggests that SOL was indeed high. His trend estimates that calories available increased until 1770 and started falling thereafter.

Floud et al (2011) did a slightly different estimate by looking at consumption rather than production. They found the calories available to be much lower, and found a steady increase in kcal consumed over time. In order to reconcile these differences, Meredith and Oxley (2013) looked at the methodologies, and reestimated the diets. They suggested that a large part of the agricultural produce could have been used as animal fodder and for seed retention, so Muldrew’s estimates were implausibly high. Nonetheless, their trend path was still closer to Muldrew’s.

With the reconciled diets data, we can reasonably conclude that diets increased up to 1770 and fell thereafter. This is actually consistent with Floud et al’s heights data which suggests that heights started falling in the late industrial revolution, where less food was available. However, the other data where diet consumption increased over time clearly contradicts Cinnirella’s estimates that heights decreased over time. The more difficult task now is to square the decrease in heights and diets in the latter IR with the increase in real wages. It might be easily dismissed that real wages were not spent entirely on food, but the heights and diets data seem to support Feinstein’s pessimistic view through a slow increase of real wages more closely.

With these proxies for SOL taken in conjunction, a weakly pessimistic view can reasonably be adopted. Economic historians tend to focus on these quantitative measures, but they should be seen together with more qualitative works. Griffin (2013) collected autobiographies and memoirs of people living during the IR and found that there was greater hope and freedom, and that some people actually enjoyed these longer working hours and apprenticeships. As such, it is helpful to go beyond the quantitative measures, and to consider how people might think that IR might have increased the SOL at that time after all. Overall, the view that there was a “rapid increase” can clearly be rejected, but a slow increase might still be reasonable.

Posted in Economics | Leave a comment

Is less state intervention better for the economy?

State intervention in the context of the economy refers to the activity of a political institution that influences the free market. The state is distinct from the government – Belgium, for instance, functioned without an elected government for some time, but the political institutions that relate to the state still functioned. Stiglitz (1987) suggested that the state is distinct from the market in that it has two features: (1) universal membership and (2) powers of compulsion that allow it to provide goods and services more efficiently than the free market, especially in the case of public goods – in other words, the state is a Coasian bargain (Demsetz), and this becomes highly evident when we look at the evolution of political institutions (North, 1991). Saying that less state intervention is better for the economy involves some extent of normative judgment, or idea of what “better” entails. Here, we will look at the main arguments for more state intervention first – the existence of market failure and the ability of the state to alleviate it, then we we will look at the main arguments for less state intervention – the possibility of government failure through imperfect information and public choice. Finally, some discussion on the type of state intervention rather than a question of “more” or “less” is also warranted. Overall, whether less state intervention is desirable largely depends on a country’s context and institutions.

Before looking at whether there should be less state intervention, it makes sense to consider why people deem state intervention as necessary in the first place, and the most common argument is market failure – the failure of the free market to be efficient. This can happen through externalities, where there are external costs and benefits accrued to third parties such that the good is either over or under consumed. One example would be positive externalities in schooling, as a more literate population correlates with higher growth and less crimes. Since the poorer people are less able to afford education, the state’s subsidy and provision of education can help to alleviate this source of market failure. Public goods will be another example – van Zandt studied the history of lighthouses and found that there was some form of state intervention in all provision of lighthouses, beyond giving property rights. Due to the nonrival and nonexcludable nature of public goods, they would either be undersupplied or not supplied by the free market, and the state can help to provide these goods. Clearly, in some contexts, state intervention is deemed necessary.

It should be noted that, for the heterodox economists, the idea of “market failure” is not very clear either, so the role of the state can be questioned. In the neoclassical school, one source of market failure is the market imperfections that arise from monopolies, where these firms produce at an output where MC=MR and charges above the socially optimum at P=MC. This results in supernormal profits for the monopolist, but a deadweight loss accrued to the rest of society. While the neoclassical school views this as a market failure, the Schumpeterian school views this as a market success! For Schumpeter, what matters is creative destruction and continuous innovation. The existence of supernormal profits for monopolists is precisely what incentivises people to innovate in order to gain market power. Thus, across different schools of thought, there is less agreement on what constitutes a market failure, and in these cases, we cannot conclusively argue that more state intervention is warranted.

Thus, the question of whether state intervention is warranted will be answered differently across different contexts and different schools of thought. Market failure through monopoly profits are less clear. Even in the context of externalities, Coase theorem suggests that when the government does not intervene, or merely assigns property rights, the free market still has the ability to correct its inefficiencies – there is some support here when Ostrom looked at the commons. In particular, for Britain in the 1700s, where commons were managed by parishes, it was rather unlikely that people could exploit the commons without the rest of the parish knowing about. Suffice to say, most people would agree that public goods are undersupplied by the free market, so state intervention is clearly warranted here. In particular, Zaire under Mobutu would have performed better with more state intervention that built public infrastructure that prevented the wearing down of public roads by 95% by the end of the military regime. As such, there are some clear cases of market failure that certainly warrant more rather than less state intervention.

There are standard arguments for less state intervention from public choice theory. Public Choice Theory suggests that we should not look at the state as an all-good and all-knowing Platonic philosopher king (Butler). Instead, we should compare a realistic market with a realistic state. (Posner) People running the government and state are themselves economic agents who want to maximise their own utility, which can come from higher budgets, or electoral outcomes. This implies that the state might not act in the interest of social efficiency (if such a term can even be defined after we have considered the market failure arguments), but act in its self-interest. This would be in the form of pork-barrel spending where politicians spend inefficiently on public goods or help concentrated interests in order to gain popularity with the swing voters. A recent example would be how Trump opted out of the Paris accord in order to protect the concentrated interests of the American workers who work in coal and steel and have practices that are not environmentally friendly, even though it takes up a very small proportion of the economy. In light of such failure, we might argue that less state intervention is desirable, so that the market can weed out its own inefficiency – as can be seen when Thatcher cut down subsidies to inefficient industries in the 1980s that eventually aided structural change and growth.

The second argument for less state intervention would be that there is imperfect information in the state. The state’s ability to provide the right subsidies and optimum amount of public goods is premised on the all-knowing nature of the state, when this is far from true. The government does not know the optimum amount of carbon emissions to cut down on – which empirically led to the price of carbon credits in Europe falling close to zero at some point. How much public goods to provide is also another political question, as Alesina noted that in US, differentials in public goods are correlated with the extent of ethnolinguistic fragmentation, suggesting that the provision of public goods came more from political decisions rather than purely economic considerations of costs and benefits to the society. In light of information gaps, mechanism design appears to be a promising way of giving the government sufficient information where each person has every incentive to reveal his true preferences such that optimal amount of provision can be achieved. One success story would be Singapore’s Certificate of Entitlement system that uses the Vickery second price auction such that people would only pay the lowest bid that made the cut for the COE. However, in most situations, such mechanisms are rather complex and difficult to implement across the state, so it appears that Hayek might be right after all – that the market knows best. Without sufficient information, less state intervention can be desirable.

It is very difficult to make any blanket judgment on whether more or less state intervention is better for the economy. The points for and against intervention mentioned above are non-exhaustive, and for each point there are many more qualifications and critiques. Perhaps, the bigger problem is that economics cannot even give a good answer for what “better for the economy” entails, for this is a normative judgment that differs across different schools of thought. Nonetheless, there can be some agreement across countries and contexts. North Korea would likely do better with less state intervention, as we have seen how the black market in Russia alleviated the inefficiency in central planning. In Africa, state intervention in the form of providing education would probably be desirable in the long run, as Glaeser (2004) notes that such development of human capital correlates most closely with growth and development. Across history, we have seen both success and failures of a strong autocratic state – successes include the East Asian countries that relied on the state for corporatism, devaluation and protectionism; failures include the communist countries that eventually collapsed. Thus, empirical evidence and theory cannot give us any conclusive answer to this question, and it is more important to consider these ideas in its context and institutions.

Posted in Economics | Leave a comment

How does religion affect economic development?

Economic development is rather difficult to define, for its definition ranges from GDP per capita to a more complex index like the Borda (Dasgupta and Weale, 1992). Broadly, we can think of development as an increase in GDP per capita with a reasonably equitable distribution, or as Amartya Sen puts it, freedom to obtain goods that are generally deemed desirable. Religion, if we take Durkheim’s textbook definition, refers to ideas that are associated with the spiritual realm. Religion can affect development through several channels including: (1) increasing savings rate, as Guiso shows that it is culturally linked (2) increasing desirable traits and prosocial behaviour, especially with regard to trust (3) influencing human capital through literacy, and (4) building social capital. We will look at each aspect in turn, and while we cannot attribute development purely to religion, religion can indeed give us insight on developmental patterns.

Social capital
Religion can help to build social capital, or create divisions. To understand this, we first need to clarify what is meant by social capital (SK). In the neoclassical growth model, g_y = a g_K + b g_L + eps, and SK can feed into this epsilon, or error term. SK is broadly defined by Putnam as horizontal associations, thought Coleman subsequently argued that it should involve vertical associations as well. Horizontal associations can be in the form of participating in group activities, interacting with others, or otherwise, and this can lead to cooperation that can be beneficial to growth – a feature of development.

A useful framework to understand religion is the Azzi-Ehrenberg model (1975), which suggests that there are three main reasons why people “consume” religion. The first reason is the salvation motive – the idea that there are benefits in the afterlife. Secondly, there is a consumption motive – that the very participation in religious activities grant them utility. Thirdly, there is a social motive – religious participation allows people to meet others who can potentially help them advance their interests in businesses or otherwise. A prime example of this occurrence is the Quakers during the Victorian period. Since they were religious non-conformists, they were unable to obtain funds from the rest of the stock exchanges, and their interactions with each other still allowed them to have successful businesses and funding. A success story would be Cadbury’s chocolate. As such, the religious structure can help to build social capital that can contribute to entrepreneurial development and growth. Furthermore, sharing is involved in many religious communities since they are organised as charities, as Buddhist temples in East Asian countries will show. Thus, religious involvement can prime an equitable distribution of resources.

Conversely, religion can work against social capital. The conflict in Northern Ireland revolved around religious differences between the Catholics and the Protestants, which led to terrorist groups and civil unrest. This subsequently led to lower income growth and development. The GDP figures relative to the rest of the UK shows this difference.

Savings rate
Religion can influence savings rate that can lead to long-run growth and development. The link between savings rate and growth is seen in the Solow Growth Model, which suggests that the steady state capital per person is given by del k = sy – dk, and as savings rate s increases, capital accumulation increases, which leads to a higher income.

Guiso et al (2004) argue that religion can influence savings rate. Using data from the World Values Survey, Guiso found that Catholics and Protestants are more likely to say that teaching thrift to their children is important than people without religion. While the Christians believe in the importance of thrift, we might question if their intent is actually realised. Since it is difficult to collect savings rate of individuals, Guiso looked at savings rates across countries, and found a correlation between their answers to the importance of thrift and their savings rate. In particular, 10 percentage points increase in the people who believe that thrift is important correlates with a 1pp increase in national savings rate. As such, to the extent that religion teaches thrift, there can be an effect on development.

This view, however, should be qualified as there could be an omitted variable bias. It could well be political and social institutions that determine savings rate rather than religion. In response, it should be noted that religion is not the sole determinant of s, but is a plausible account amid other factors.

Increasing desirable traits
World religions are generally associated with morality. Tabellini notes that higher levels of trust, reliance on effort, and general morality do correlate with growth. Now, we need to establish that religion does indeed result in more desirable traits, and that such traits correlate with growth.

Religion resulting in prosocial behaviour can be seen in Norenzayan and Shariff’s (2008) experiment. They placed participants in trust games, and found that when religious ideas were evoked, participants tend to cooperate more and act more favourably towards others. In particular, there are higher levels of trust. This is what they call “supernatural monitoring”, as believing that God watches them helps to increase prosocial behaviour.

Trust levels correlate to growth. Arrow notes that every economic transaction involves a certain degree of trust, so it is a prerequisite for economic activity and consequently growth. Using the World Values Survey, Guiso found that there was indeed a correlation between trust and growth rates. To explain this mechanism, Knack and Keefer suggest that in the absence of formal institutions, trust was important in developing countries for property rights and economic transactions so that economic activity can occur. In other words, trust worked in place of stable political institutions, which gives us a plausible account for how religion aids economic development.

Improving human capital
Human capital (HK) is distinct from social capital in that HK is determined by individuals while SK is determined by associations between individuals. HK would be in the form of ability, primarily measured by literacy rates and education levels. Studies on this aspect are seen in the works of Weber, Myersson, and Majid.

Weber is probably one of the first social scientists to relate religion to development, and this began with his observation between Catholics and Protestants. Weber noted that Protestants had a better work ethic as their beliefs involved pleasing God through hard work in secular lives rather than pure ascetic living. Iannaccone supports this argument by noting the influence of Protestantism on literacy rates: when the Protestants wanted the ability to read the Bible for themselves, literacy rates in Protestant dominated towns increased. Historical records also showed how the protestant reformer Martin Luther developed schools so that people would be literate enough to study the scriptures for themselves in their native language – this also benefited women’s education. Additionally, Rubin (2014) notes that by 1500, Protestant towns are more likely to have a printing press than Catholic towns, and the result is robust even when considering the intensity of printing.

However, we should note that the type of beliefs adopted in Protestantism also matters significantly. Barro and McCleary (2003) found that greater beliefs in hell results in higher growth rates and incomes, and this can be attributed to hard work. In contrast, higher church attendance correlates with lower growth – we should not be surprised to see this result as church attendance causes the diversion of resources from the rest of the economy.

HK is also affected by politics when religion is dominant. Myersson (2014) found that when a conservative Islamic leader was elected in Turkey, there was higher secular educational attainment for women. While the mechanism is not very well understood, it is possible that conservative leadership led to the construction of more female dormitories, which result in conservative families being more disposed towards their daughters obtaining higher education.

Finally, religious practices can affect human capital. Majid found that when mothers fasted during Ramadan during their first month of pregnancy, children are likely to have lower birth weights and score lower on IQ tests. Other research also found that adults are 20% more likely to be mentally disabled when their mother fasted during pregnancy. This is consistent with the Barker hypothesis which suggests that life expectancy at 50 depends on the month of birth, as people whose mothers were pregnant during winter where little food was available tended to have lower life expectancy.

It should be noted that human capital is one of the strongest proxies for differences in development. Glaeser (2004) argued that there is a strong correlation between growth and the level of education. This is not surprising as we should expect people who are better educated to go into value added sectors and generate higher incomes.

The four main channels that religion could affect economic development are through the savings rate, human capital, social capital, and morality. There are other effects including institutional effects, as Kuran (2004) compared the Islamic waqf with the western financial system to account for growth differentials, so the suggestions above are clearly non-exhaustive. Iyer (2016) notes that most of the work that relates economics to religion are done in Christianity and Islam, and clearly more studies are warranted in Hinduism and Buddhism, which are other dominant religions in the world. This remains an interesting field of study and when we look at the sociological backdrop of economic activity, we can gain greater insights on development.

Posted in Economics | Leave a comment

What’s missing in the Eurozone?

We had one lecture today on European Integration by Max Beber. I didn’t intend to attempt the question on European Integration during the examination, but thought it would be interesting to attend the lecture anyway – and I was not disappointed. The main question he addressed was about what was missing in Europe’s economic and monetary union (EMU).

Beber began with the theoretical origin of EMU, and this began in the 1980s. Most economies were facing stagflation problems (high inflation and high unemployment) in the 1970s, possibly as a result of their demand management stop-go policies in the post-war period. In the 1980s, there was recognition among academics that the Keynesian paradigm of demand management was no longer tenable, and the monetarist paradigm seemed attractive, thereby signalling the transition to the neoliberal period. Friedman suggested that the long run Phillips curve was vertical, so unemployment and income were fixed, and demand management policies only resulted in the upward shift of short-run Phillips curves. If demand management policies (such as fiscal and monetary policies) are no longer effective, the economists were more open to other policy options. Europe was also getting more integrated at that time, so abandoning the individual countries’ own fiscal and monetary policies for an EMU seemed like a viable plan.

We can put this into a cost-benefit model. If we plot integration on the x axis (where “integration” refers to the amount of integration required for EMU to be feasible) and cost/benefits on the y axis, we will have a rather counter-intuitive result. The benefit curve will be upward-sloping: since the whole idea of an EMU is to reduce transaction costs, more integration means more benefits. The cost curve will be downward-sloping: when economies are more integrated, they are less prone to asymmetric shocks and the opportunity cost of holding their own currency for monetary adjustment becomes lower. Beber argues that the cost curve for the monetarist paradigm lies to the southwest of C_Keynesian. This means that, at least perceptually, less integration was required for EMU to be viable, and it seemed attractive in the 1980s.

Beber’s historical account on the intellectual origins of the EMU based on the arguments of the 1980s is probably the third account of EMU origins that I’ve heard in Cambridge. Of course, they are not mutually exclusive, but it is interesting to hear what different people have to say. The first account I heard was by my supervisor. Kuczynski noted that the stagflation problems in Europe was due to a bifurcation of policies in the 1960s. After the supply of cheap labour dried up from Mediterranean Europe, and the post-war growth has slowed (if we refer to the Solow model), economies had slow growth, and rising prices. While Europe was already integrated by trade, the German block appreciated their currency (since they were terrified of the inflation they once faced in the 1920s), so that they could have low inflation and low growth. The non-German block (like France and Italy) depreciated their currency, so that they could have more growth, at the cost of more inflation. Since the economies were already integrated, Europe had the worse of both outcomes – slow growth and high inflation. This led them to scramble for a solution through the EMU. The second account I heard was by Nigel Knight, our lecturer for Governing Britain. After the war, philosophers were contemplating on the causes of war, and concluded that the major cause of war was nationalism. In order to overcome nationalism, they needed a supra-national block, and this had to be built gradually, so it began through trade in the immediate post-war through the European Coal and Steel Community. This would later result in deeper integration, hoping eventually to form a political union to overcome wars, so EMU was just a natural step towards that. I’m not taking sides, but I think it is quite interesting to hear their non-mutually exclusive accounts.

With the origins in mind, Beber looks at the characteristics of regions in general. Firstly, there could be differences in relative competitiveness: when German labour productivity increases faster and unions are more sensible than Spanish ones, German goods tend to be produced cheaper, and Germany gains in relative competitiveness. Secondly, there could be asymmetric shocks: if there is a drought in Greece that destroys all the olive crops, Greece would be affected more than other countries. With differences in relative competitiveness and asymmetric shocks, external imbalances will arise, such that there is more economic activity in some areas than others. In particular, if a region imports more than it exports, then NX is negative, and this has to be financed. Financing external imbalances can only occur in the short run, and in the long run, adjustment must occur. Adjustment can come in two forms (1) price differentials (2) factor mobility. In the example between Germany and Greece, we would then expect a depreciation of German real exchange rates.

After we unpack the whole mechanism behind external imbalances, it becomes very easy for critics to point out why EMU was a bad idea. Before we go into that, we should consider, how did the EMU architects convince us that EMU would work? Starting from the external imbalances, they argue that the relative competitiveness will be evened out eventually. There are policies such as the cohesion funds, and the attempt to invest in research and development and spread best practices around Europe that can even out relative competitiveness. The arguments for the removal of asymmetric shocks are very interesting. In the Keynesian paradigm, it is argued that demand shocks are caused by changes in private investment. But in the monetarist paradigm, demand shocks are caused by policies. This means that once countries abandon fiscal and monetary policies in favour of EMU, these shocks would not exist in the first place. As for exogenous shocks, I read in Mankiw and Taylor that some argue that integration is endogenous – when the EMU project began, it gave businesses confidence to invest in other countries, so integration will increase as the EMU project deepens. With greater integration, there will be less asymmetric shocks that cause specific countries to suffer immensely. In this light, the problem of external imbalances would not be pressing. Even if there is a problem, the common currency would make financing easier in the short run, because of currency stability. In the long run, the channel of price differentials through exchange rates has been removed, but there is still factor mobility. Tariff barriers have been removed from EU long ago, and Europe continues to lower non-tariff barriers, and allow greater mobility of labour and capital.

Now, we move on to the critics. Kuczynski noted that greater integration, in fact, could worsen asymmetric shocks (noting that coincidence of business cycle is the McKinon criterion for EMU). Demand and supply shocks usually occur by industries and sectors. As the economies become more integrated, they tend to specialise in the area in which they have comparative advantage, as Ricardo’s trade theory goes. This means that if there are shocks to specific industries, a particular economy will be affected more than if the integration had not occurred in the first place. Thus, even if we accept endogenous integration, the McKinon criterion has yet to be fulfilled. The Mundell criterion for integration – factor mobility, is also unfulfilled completely in the EMU. There are still “health and safety standards” between countries, which effectively function as non-tariff barriers.

My favourite critique would probably be Friedman’s daylight saving adjustment argument (which I just learned today!). If we want to maximise daylight, there are two ways to do it: (1) we could change all the timetables of schools, trains, offices, shops etc. or (2) we could shift the clock back by an hour. In light of the external imbalances in the economy, adjusting through the exchange rate is very much like shifting the clock back – a simple mechanism that solves the problem. It seems, instead, that the EMU architects opted for option (1).

If we look back at the mechanism behind external imbalances again, it is clear that there is one crucial point that is missing that would allow the entire system to adjust – fiscal transfers. Regional imbalances occur all around the world, and fiscal transfers do take place, where money moves from London to Northern Ireland, or from North to South Italy. These transfers do not create a problem because the Londoners do not expect the northern Irish to pay them back, but Germany would expect Greece to pay them back, if loans were made. Considering how the EU Budget is only 1% of EU’s GDP, Brussels is far from functioning like a central government that can easily make fiscal transfers. It is also notable that 80% of the budget is tied up in the Common Agricultural Policy.

Considering the inadequacy of the fiscal state in Europe now, it seems logical to just persuade the Eurozone to have a larger budget, or to push for a fiscal union. The problem, however, is that fiscal budgets are politically tied, and it is difficult to have a fiscal union without having a political union. The latter, it seems, is still far beyond what the Europeans are comfortable with. [I hope there are some Euro-barometer statistics to back this up.] If Knight is right about the final intent on there being a political union, then EMU might not be such a bad project after all.

Meanwhile, Beber did point out a rather roundabout way of dealing with the fiscal problem, and that is through a “banking union”. If you read the fine print, the banking union effectively requires losses of defaults to be spread across the entire system. If there are defaults, the borrowers gain, and the lenders lose out. This, then, indirectly transfers wealth from the lenders to the borrowers. Lenders are generally the people in wealthier (more competitive) nations, and borrowers the people in the poorer (less competitive) nations. [If intuition fails, we can use equations: NX = S – I, so if NX>0, S>I] Thus, effectively, there are still redistributive effects of wealth.

If a conclusion might be drawn (or just have a reasonable answer to the question in the title), I’ll say that a fiscal union is missing in the EMU project, and this is inherently tied to politics. Okay, so I’ve done a rather lengthy article, but I’m still not confident on doing the European Integration question during the exam, possibly because I don’t think I know enough, or I have a comparative advantage here. Well, at least I had fun writing this.

Posted in Economics | Leave a comment

Readings and the mediocre essays

Evidently I have not been keeping up my writing since Tuesday. There are two main reasons (1) I’m beginning to question the efficiency of this pedagogical tool (2) I found other things to do that seemed to be a lot more efficient. Here, I’ll suppose efficiency just refers to the increase in the expected number of marks for every unit of effort put in to study (as if either is measurable at this stage).

From Wednesday onwards, I began to question the efficiency of writing blog posts. Writing based on current knowledge does help to fossilise one’s knowledge and to help one think, but this is contingent on the individual already possessing sufficient knowledge and facts for this to be meaningful. Initially, I thought I knew enough to start this process of consolidation and writing, but I soon found out the limitations of my knowledge when I started reading later in the week.

There were at least two posts that I had previously written where I felt like what I said was inadequate in light of the new readings that I had done. The first post is one the power structures in the economy. On Thursday, I read this interesting book by Lukes (1974) talking about a radical view on power. He suggests that there should rightly be three dimensions. The first dimension is adopted by the pluralists such as Dahl, who believe that A has power over B when A is able to influence B to do something that B would otherwise not do. Bachrach and Baratz (1970) accepted the first dimension and added on a second dimension to power, which is the ability to suppress conflicts, citing Baltimore as an example where the institutions deflected the views of the blacks such that it was difficult for them to air their public concerns. Lukes argued that the two dimensional view was limited and that a third dimension was necessary – this addressed not just the suppression of existing conflicts, it also suppresses latent conflicts. This explains how institutions and inactivity could keep their views suppressed for so long. In this light, the previous post on power relationships in the economy would be deemed by Lukes as largely one dimensional as it only looks at how people in authority influence the decisions of others. This reminds me of the Flinders reading (2012) where the market system uses money to influence not just behaviour, but also thought. Anyway, the second reading that changed my perspective slightly was Eichengreen’s article (2004) on the interwar economy. In particular, he talked about how protectionism would result in the appreciation of the exchange rate since lower imports must be balanced by lower exports. He qualified that exchange rate would not increase massively because Britain operated under a managed float system. This explanation was actually right in my face when I studied the Mundell-Fleming model in macroeconomics, but I guess i hadn’t thought about it.

My existing knowledge of the topics are largely confined to the lectures that I had sat through and the small amount of readings that were done. From the examiner reports, it appears that such essays would not do well, so sticking to this approach might not give the desired returns in the exam. Perhaps, I should drop this project and go for the readings instead.

While Wednesday and Saturday were spent largely on drilling microeconomics, mathematics, and statistics, I did have a fair bit of time to do readings on Thursday and Friday. I learned about what Skidelsky thought about the fall of Keynesianism in post-war Britain. Rather than blaming the stop-go policies all the time, a more qualified viewpoint is required as unemployment never went above 3%. There were theoretical underpinnings such as wage push inflation being the primary cause of inflation at full employment such that the government adopted corporatism in the post-war, which eventually worked against them through strikes. There is also the possibility of a long Kondratieff cycle in the post war at work rather than the Keynesian policies adopted. It was also interesting to read about the role of the state as discussed by Stiglitz. He suggests that there are two main characteristics of the state that allowed it to act differently from the market: the first is universal membership, and the second is the powers of compulsion. These can then allow greater efficiency in providing goods such as unemployment insurance and pensions. Medema’s survey of the history of public choice theory was also quite interesting, though the only part that I think would be vaguely useful for essays is Arrow’s impossibility theorem in creating an ideal voting system. I feel like I’m learning more through these readings than just drilling.

Time went by so quickly last week. Today’s a good day to keep the Sabbath, and to rethink my study plans towards the exam.

Posted in Uncategorized | Leave a comment

1931-2 policies a consistent policy framework?

Do the set of policies introduced during 1931-32 constitute a consistent policy framework for the British economy during the 1930s?

The set of policies introduced in 1931-32 include departure from the gold standard (1931), the general tariff (1932), expansionary monetary policy (1932), and slightly contractionary fiscal policy. Consistency as a policy framework requires that the policies do not have contradictory outcomes. Each policy has macroeconomic implications, so this essay will look at each policy set in turn, and evaluate if they can be collective coherent. Overall, while there are some interesting points for discussion, the policy framework can be regarded as largely coherent.

Leaving the gold standard
When discussing the departure from the gold standard in 1931, it is interesting to consider if Britain actually “left” the gold standard. Britain’s currency was overvalued since it entered the gold standard at pre-war parity, so there was a downward pressure for the sterling to depreciate, and this came at the cost of lower exports and growth. IN 1931, the pound came under speculative attack, and the risk premium would feed into higher interest rates that leads to further pessimism. Thus, it was likely that the pound would exit the system anyway (as it did from the ERM in 1992), so this departure could be seen as a consequence of the market rather than a “policy” per se. However, it is notable that the government had a choice – the pound’s value can still be defended on the international market if the government were to increase foreign debt to buy up excess pounds. It was effectively a choice to allow the pound to depreciate rather than increase Britain’s debt further, so it can still be considered a “policy”.

Leaving the gold standard, in principle, shifted Britain’s position in the policy trilemma. The policy trilemma suggests that countries can only choose at most two out of the following three objectives: (1) fixed exchange rates (2) monetary independence (3) free capital flows. Britain chose 1 and 3 from 1925-31, and the position shifted after giving up fixed exchange rates. The immediate effect of leaving the gold standard was the sterling’s depreciation, but Britain’s policy regime in the 1930s was not completely floating either. It was a form of managed float system, which makes theoretical analysis more challenging. Nonetheless, this depreciation allowed Britain to increase net exports (NX) and hence recover, assuming that the Marshall Lerner condition holds.

Expansionary monetary policy
After leaving the gold standard, Britain waited for a year before beginning its expansionary monetary policy regime. This cautious approach was largely due to the inflation that many of its neighbours faced soon after depreciation. Expansionary monetary policy was implemented through a lower interest rate, which will increase interest sensitive consumption and investment, thereby aiding Britain’s recovery.

Expansionary monetary policy is consistent with the exchange rate regime, since moving away from the gold standard gave Britain monetary independence. It should be noted, however, that monetary policy is inconsistent with conscious depreciation. In a floating exchange rate regime, output Y is determined in the money market, and the IS curve and exchange rate adjust to the fixed Y. Since Y is fixed by the LM curve, lower e will just be absorbed by higher r, due to capital outflows from a cheaper sterling, which will cause a fall in investment, so Y still remains. Nonetheless, if we see depreciation as a result of leaving the gold standard rather than through conscious depreciation, it is still perfectly consistent with expansionary monetary policy.

General Tariff
Britain left its unilateral free trade position in 1932 when it adopted the General Tariff, where industries had a 10% tariff protection rate on average, though some industries were protected more and others less. The General Tariff, in theory, will lower imports. This is meant to increase NX and hence help with Britain’s growth.

Under flexible exchange rates, Y is determined by the money market. If the money supply does not change and Y remains stagnant, the tariff would not change Y. Tariffs would just lower imports, which must be balanced by lower exports, so the exchange rate will appreciate. This, clearly, did not happen in the British economy as the money supply was increased through expansionary monetary policy, so that the fall in M can increase NX and Y, with money supply to back it up. This further emphasises the consistency of this policy package.

Contractionary fiscal policy
The policy stance for the government budget was slightly contractionary because they utilised a balanced budget, but unemployment rate was still high in the 1930s.

On first glance, a contractionary policy might run contradictory to the existing expansionary policies. However, two qualifications should be made (1) fiscal policy was not a point of emphasis in the 1930s, and was not intended to be a tool for economic recover, but to put the government on track (2) contractionary fiscal policy is not entirely inconsistent with expansionary monetary policy as an increase in government savings shifts the savings curve outwards to lower interest rate, which is consistent with expansionary monetary policy. Furthermore, since monetary policy works better under flexible exchange rates than does fiscal policy, the government was wise not to use fiscal policy as a tool for recovery.

External consistency
Thus far, we have discussed the internal consistency of the policies within their respective frameworks, and showed that this particular policy package was theoretically sound. It would be interesting, then, to consider if there is external consistency – the idea that the policy package met the needs of the economy at that time, and that empirical results were favourable.

The policy package was clearly consistent with the economy’s needs. Britain was recovering from the 1930 recession, so expansionary policies to boost AD and Y were definitely important.

Scholars have also shown empirical evidence on the success of Britain’s policies. Broadberry and Crafts, using a difference in difference methodology with a three way classification, note that additional tariffs did benefit industries. Kitson and Solomou also showed that newly protected industries in Britain benefited from the general tariff. Eichengreen and Sachs did a study of 10 countries and found that those that left the gold standard performed better than countries that did not. While results are rather mixed, depending on the perspective adopted, we do have sufficient evidence to conclude that there was some success in these policies, and Britain’s relatively fast recovery compared to the rest of the world is a sign of that.

The policy set adopted in 1931-2 was a consistent framework for the British economy during the 1930s, as they worked together to stimulate Britain’s recovery. There are interesting combinations of policies like tariffs on a floating exchange rate system, fiscal and monetary policies going in different directions, and devaluation accompanying a monetary regime, but if we were to analyse these policies carefully, we will find that they are still broadly consistent.

Posted in Economics | Leave a comment