By Peter H. Lindert, Cambridge University Press, April 8, 2021, 1108478166

Peter Lindert worked with and is friends with Alan Olmstead. I’ve met Peter a number of times. I don’t think he mentioned the book, but his daughter, Kathy did.

My biggest take away is that social spending works, but we have to be careful to spend it on the young, not the old. Too many countries have fat pensions that give social spending a bad name.

Making Social Spending Work is full of important statistics and details. Lindert knows his stuff. I am no one to judge, but it is very interesting despite heftiness and myriad detail.

[k401] The debate has raged for at least eight centuries. The words have changed, but the opposing positions have not.

[k403] Each side of the debate sends a respectable signal along with ideological noise.

[k406] For centuries now, the core argument against tax-based social spending, or all civilian government spending, has warned of perverse incentives. The incentive argument has economic plausibility and deserves careful testing.

[k420] All of these respectable criticisms deserve careful testing.

[k423] The noise includes demonizing images of greedy and insensitive billionaires, robber barons, and corporations.

[k425] The respectable signals strike notes that are replayed in today’s economics textbooks. For basic welfare spending – that is, assistance to poor families and unemployment compensation, the incentives critique is not rejected altogether. Rather, it is rebutted by denying that poor people lost work by choice, or that working more is a valid option without government help.

[k450] More fundamentally, the conflict of self-interests is what generates controversy forever.

[k467] Finding #1. A country’s government social spending takes off only after the country has both the fiscal capacity and the political will to build safety nets.

[k479] Finding #2. A major barrier to growth and equality has been the refusal of those with power to devote taxes to universal education.

[k499] Finding #3. Since around 1910, there has been a long mission shift in social spending, toward support for the powerful and elderly, at the expense of assisting the young and the poor.

[k514] Finding #4. Larger social-spending budgets have not produced any net loss of GDP, or in skills, or in work.

[k535] Finding #5. Governments around the world have shifted toward equalizing incomes since 1910.

[k555] Finding #6. A short-run threat to the future of social spending looms from waves of refugees and the nationalist reactions to these waves.

[k785] The city of Ypres, after seeking the advice of the renowned writer Juan Luis Vives, assigned to four dischmeesters (distributors of alms) in every parish the task of investigating the state of the poor, their occupations, age, number of children, illnesses, income, whether they were well behaved, and whether they were drunkards or beggars. Ypres thus may deserve the title of pioneer in data-based social monitoring of the poor.

[k1366] In 1872, according to a Parliamentary Report by the Local Government Board, a family on indoor relief cost the taxpayers 10 shillings a week, whereas a family allowed to receive outdoor relief cost them only 4 shillings a week.

[k1427] Particularly popular was their Old Age Pension Act, passed in

  1. It was non-contributory, to be financed by the taxpayers rather than the workers’ earlier contributions from their earnings. Compulsory health insurance followed in 1911.

[k1547] The Dutch lead in literacy began early, helped by the pre-Reformation foundation of the Brethren of Common Life (BCL) by Geert Groote in 1374.

[k1575] Indeed, leading thinkers at the time had already come down on the side of tax-based schools on the grounds of externalities. Consider what Adam Smith, Robert Malthus, and Thomas Jefferson wrote about tax-based mass public schooling even before it happened. Smith felt that a society at large should be ready to pay taxes for education’s externalities, the part of its societal benefits that were not captured by the private providers or the students: “The expence of the institutions for education and religious instruction is [like the expense of maintaining good roads and communications] beneficial to the whole society, and may, therefore, without injustice, be defrayed by the general contribution of the whole society.”

[k1594] Thomas Jefferson held a similar view of tax-based schooling. In 1779, Jefferson introduced his Bill for the More General Diffusion of Knowledge in the Virginia assembly, calling for a statewide system of free public elementary schools to be paid for by local taxpayers.

[k1601] Yet each time he introduced a taxes-for-education bill in Virginia – in 1779, in the 1790s, and again in 1817 – it was defeated by those whose self-interest would be compromised by property taxes that would pay for common schools.

[k2109] Sources and notes: See Figure 5.1 That American lead in average years of education was sustained impressively until some time around the middle of the twentieth century, by which time the Americans had seized a particularly wide lead in secondary education (Goldin 1998, 2001). Since then, other countries have been catching up, though none of them has yet clearly overtaken the United States in this respect.

[k2118] Russia, Japan, and Korea have all been slowly converging toward the American schooling average for decades. India remains far behind, as does China.

[k2234] We are indeed getting smarter. The global trend toward better IQ scores, the “Flynn effect,” points to an average gain of 2.8 IQ points per decade, or seven points (almost half a standard deviation) per twenty-five-year generation. The improvement is common to peoples taking the IQ tests in all thirty-one countries.

[k2245] One glaring result illuminates the controversy over race and IQ. Flynn (2008, 2012, pp. 135–140) finds that blacks of a given age group in 2002 had IQs slightly higher than whites in 1947–1948, and far higher than whites in 1909. This result, like many others, implies that environmental factors account for IQ trends much more strongly than any possible factors inherited at birth.

[k2503] Friedman’s key emphasis was on the first two institutions listed above: (1) equal funding per student, and (2) free parental choice.

[k2507] He explicitly confronted the fact that something like a voucher system was already being proposed by segregationists in the American South, who sought local support to keep blacks out of subsidized white schools. He grappled unpersuasively with this problem, and in the end retreated to the pious hope that free-market competition would somehow defeat racism.

[k2512] As Charles Tiebout (1956) theorized, people in major urban areas can choose their governments at the local level. Some prefer higher taxes that pay for better schools and other local public goods, while others do not. Thus, in a major metropolitan area, with many suburbs to choose from, people can vote with their feet – if the cost of moving or of commuting to a nearby school is zero.

[k2523] The devil is in the details, as Friedman himself seemed to anticipate. The history of school choice shows that the effectiveness of subsidized school choice in promoting efficiency and equity depends critically on certain institutional constraints on educators’ freedoms. Even the Dutch, Belgians, and French – our pioneers in subsidized school choice from the early twentieth century – evolved a set of constraints on the process of matching schools with students. All three of them reinforced the effect of learning efficiency on parental choice by instituting high-stakes centralized exams. And typically, as we noted in connection with Dutch practice, publicly funded private schools cannot “top up” the cost of their schooling with extra fees, nor are they allowed to engage in selective admissions.

[k2799] The 1967–1980 test-score decline was apparently caused by something that happened to American children after fourth grade.

[k2802] The first is the sudden consolidation of America’s school districts, which seriously restricted parents’ school choice. Between 1940 and 1970, the number of separate public-school districts plunged about 85 percent, while the number of students soared.

[k2805] The giants, led by the City School District of the City of New York (over a million students) and the Los Angeles Unified School District, set up more rigid rules blocking parents’ choices.

[k2809] That second force is the spread of state legislation facilitating the rise of teachers’ unions. The period from 1960 to 1974 saw a jump in the share of states where laws facilitated collective bargaining and permitted teachers’ unions to strike under certain conditions. By 1988, when only 19 percent of the US labor force was unionized, fully 75 percent of public-school teachers belonged to unions.

[k2818] A third force that seems to have weakened the delivery of quality learning in American schools also took the form of a restriction on choice. A few states passed laws imposing funding-equalization formulas, in response to various complaints about unequal school funding.

[k2826] Unless similar things silently happened in other OECD countries, with similar timing – as seems unlikely – it would appear that these three choice-restricting developments in the United States since the 1960s have lowered its average learning relative to other countries.

[k2859] More and more, the growing social budgets drifted away from education and from helping the poor, toward insuring the powerful and elderly against outliving their retirement nest eggs.

[k2862] “Social insurance has increased far more than public assistance, so ‘rise in the social insurance state’ is a far better description of the century than ‘rise in the welfare state.’”

[k2954] The positive impact of women’s voting power was especially evident in the public expenditures on education and health.

[k3601] Why would they spend so much more on each person older than 65, relative to what they spend to educate each person in the 5–21 age group? Their

[k3604] Rather, these countries’ pensions stand out for being funneled toward those in the formal sectors who had high lifetime earnings, and especially toward those in the better-paid ranks of the public sector, such as top executives, military officers, judges, and teachers.

[k3611] Why do Turkey and Brazil have an average pension payout that is greater than half of GDP per person in the working-age group, ages 18 to 64?5

[k3629] Social spending in these countries tilts not only away from education but also away from the poor, according to recent OECD comparisons.

[k3633] Brazil may have the region’s most inegalitarian and anti-growth pension regime. Regressive pensions are cemented into Brazilian law by the 1988 Constitution.

[k3672] There are no known growth benefits to be had by taking money from the general taxpaying population and giving it to the current elderly, whereas if the same funding had been channeled into education, especially primary and secondary education, it would have reaped the high rates of return documented back in Chapter 5.

[k3676] As one can make out from Table 7.1 and Figure 7.3, countries further from the equator lean toward supporting the young, in this case by shifting tax money to education. Most, but not all, of these higher-latitude countries tend also to be rich democracies.

[k4248] There has been no clearly negative correlation between social spending and GDP per capita anytime since the dawn of widespread social insurance in the 1880s.

[k4963] Why should the four low-budget free-market countries of North America and Australasia be the laggards in adopting the VAT and other indirect taxes, as shown in Table 9.1 and Figure 9.1? In particular, why should the United States have zero VAT at the federal level, and only about 8 percent of all tax revenue even if one included state sales taxes as if they were a VAT?

[k4969] As Murray Weidenbaum, a former chair of President Reagan’s Council of Economic Advisers, said: “Many analysts believe that it is fairer to tax people on what they take from society, rather than on what they contribute by working and investing.”

[k5123] The third pro-growth feature of the welfare states’ social spending has been their relatively generous investments in career continuity and skills accumulation for mothers, indirectly also supporting their children’s early development.

[k5239] At this bottom end of the spectrum, the United States has attained some notoriety for being the only country, other than Papua New Guinea, not to provide any national government payments for parental leave.

[k5626] The twentieth century delivered a nearly global shift toward progressive (from-rich-to-poor) redistribution, even during the infamous rise in the inequality of market incomes since the 1970s.

[k5630] As a corollary, the rise in income inequality since the 1970s owes nothing to any widespread retreat from progressive government policies, because no such retreat happened. On the contrary, households’ net incomes became increasingly unequal despite a tendency of government to equalize their resources.

[k5636] The usual one-year measures of redistribution between richer and poorer groups do not capture the effects on lifetime income inequality, the inequality we really care about most.

[k6156] A key implication of the mass-education effect will be that we have understated how much public education spending has cut income inequality over the last 150 years.

[k6435] So the answer to the net-burden question is clearly “no,” not a net burden for any year beyond about the sixteenth year after arrival. The long-run fiscal effect of extra immigration is clearly positive.

[k6716] First, it is unlikely that transfers to your grandparents, progressive as they may be, are the most productive use of society’s government revenues.

[k6719] Second, the current elderly generation is as wasteful of the environmental future as any generation in human history – surely as wasteful in absolute terms, and probably as wasteful per capita.

[k6730] Its results show that only for the most educated groups of workers (Ph.D.s and professional degrees) did federal workers get paid less overall. For this top-education group, any extra old-age benefits were outweighed by a lower value of straight pay, canceling any gain in total compensation. In sharp contrast, federal workers with no college education got more than their qualifications would have been paid in the private sector, either in straight pay or (especially) in benefits, giving them more total compensation.

[k7121] No developed country has a greater pension problem than does Greece.

[k8273] Even where Japan’s social legislation seems to have lent a hand to women’s careers better than in other countries, as in the parental leave policies noted in Chapter 9, little change seems to have resulted.

[k8312] The divergence is caused solely by the fact that Japan is aging so much faster than the United States, due to Japan’s rock-bottom birth rates and immigration rates.

[k8870] The public’s ratings are backed up, indirectly, by the well-known survival results: for all that extra health spending, Americans die younger. The public’s ratings are backed up, indirectly, by the well-known survival results: for all that extra health spending, Americans die younger.

[k8886] While the whole sector can be called “transparency challenged,” American hospitals are particularly opaque about prices.

[k8895] The need for government regulation of health-sector pricing is globally recognized, although least so in the United States.

[k8978] In December 2019, the American president and some state governors unplugged part of this automatic stabilizer by denying even food stamps to impoverished childless adults. The loss of food stamp options left them totally exposed to the job-market collapse of 2020, without a reliable food source.

[k8994] Since children make up one third of the American poverty population, what happens to them should loom large in any debate over safety-net spending.

[k9002] Bailey and co-authors are now able to trace 17.5 million Americans living in the 2000–2013 era back to their births, or conceptions, thanks to census-type data on their dates and places of birth.

[k9013] The giant sample finds so many significantly positive effects on their adult wellbeing that the Food Stamp program since 1961 has clearly “paid for itself” in the long run, in the broad sense of raising each child’s later adult income, quality of life, and life expectancy, while also improving social externalities. Intriguingly, even though the food stamp applies to all age groups, its positive effects stemmed mainly from exposure to food stamp availability in early childhood, and even between conception and birth.