Austerity Policies and Mortality in Spain After the Financial Crisis of 2008

Abstract

Objectives. To analyze mortality in Spain and the United States before and after these countries implemented divergent policies in response to the financial crisis of 2008.

Methods. We examined mortality statistics in both countries in the years 2000 to 2015. Spain started austerity policies in 2010. We compared differences in mortality ratios, on the basis of trends and effect size analysis.

Results. During 2000 to 2010, overall mortality rates (r = 0.98; P < .001; Cohen’s d = −0.228) decreased in both countries. In 2011, this trend changed abruptly in Spain, where observed mortality surpassed expected mortality by 29% in 2011 and by 41% in 2015. By contrast, observed mortality surpassed expected mortality in the United States by only 8% in 2015. As the mortality statistics diverged, the effect size greatly increased (d = 7.531). During this 5-year period, there were 505 559 more deaths in Spain than the expected number, while in the United States the difference was 431 501 more deaths despite the 7-fold larger population in the United States compared with Spain.

Conclusions. The marked excess mortality in 2011 to 2015 in Spain is attributable to austerity policies.

Poverty continues to be one of the main causes of mortality, even in the wealthiest countries.¹ Governmental policies at the national level set the course toward social prosperity or decay, and closely linked indicators such as population-level mortality and life expectancy.² Although health systems are the best political tool for the recovery of lost health, the fight against poverty outstrips the capacity of these systems.³ Policies aimed at wealth redistribution and well-being can reduce poverty, making it feasible for health systems to decrease persisting health inequities among social classes.^4,5

Toward the end of the first decade of the 21st century, the huge financial crisis that erupted in 2008 in the United States and Europe led to the failure of many banks and immediately resulted in rising unemployment and very high levels of employment precariousness.⁶ The policies implemented on either side of the Atlantic to deal with this crisis were divergent. In the United States, measures to manage the crisis were put in place early in 2009 through policies intended to favor economic expansion by increasing public investment, salaries, and social spending.⁷ In Spain, the year 2009 saw an attempt to emulate policies implemented in the United States, but this was followed soon afterward by a radical shift during the second half of 2010 in favor of austerity policies⁶ consisting of restrictions on public investment, salary reductions, and cutbacks in social spending.⁸

The consequences of these widely contrasting policies have likewise differed markedly. In the United States, unemployment rates have fallen to very low levels,⁹ whereas in Spain, unemployment reached the highest rates in recent history.¹⁰ The economy recovered promptly in the United States, with gains in the minimum wage.⁷ By contrast, the Spanish economy has taken much longer to recover, and the outcomes of this process have included lower wages than before the crisis along with greater poverty among workers.¹¹

Although the negative consequences for health in the Spanish population have been reported,^6,12 the repercussions of austerity policies on mortality indexes have not been investigated in depth for overall mortality or mortality from the most prevalent causes, nor have mortality rates in Spain been compared with those in other countries that have implemented different economic policies. The aim of this study was to compare the evolution of overall mortality and mortality from the leading causes in Spain and the United States since the end of the 20th century, before and after the implementation of divergent policies in response to the economic and financial crisis.

METHODS

We based our analysis on mortality statistics published by governmental institutions in Spain and the United States between 2000 and 2015. In Spain, the data source was the Instituto Nacional de Estadística (National Statistics Institute),¹³ and data for the United States were from the Centers for Disease Control and Prevention, National Center for Health Statistics.¹⁴

We obtained data from these sources for the resident population in Spain as of July 1 of each year, the number of recorded deaths, age-adjusted overall mortality rates (AAMRs), mortality rates from suicide, and mortality rates from cardiovascular, oncologic, and respiratory diseases, all adjusted for age.

Population and mortality data are reported here as absolute frequencies and mortality rates expressed per 100 000 inhabitants. We estimated adjusted deaths from the AAMR for the general population in each year. We then calculated the observed mortality rate ratio (OMRR) as AAMR for each year divided by AAMR for the year 2000.

Austerity policies were implemented in Spain in the second half of 2010, so we considered mortality rates up to and including 2010 as the pre-exposure period, and rates from 2011 on as the exposure period. To calculate changes in mortality trends and excess deaths related to the exposure required assessment of time trends and estimation of the expected number of deaths (that is, what would have happened if the austerity policies had not been implemented). So, we obtained a linear trend for the evolution of OMRRs from 2000 to 2010 by including the calendar year in the models; similar procedures have been used previously to study mortality trends.^15,16 Assuming the same trend for the period from 2011 to 2015 as observed in the preceding decade, we estimated the expected mortality rate ratio (EMRR) and its 99% confidence interval (CI) with the same equation as for the theoretical trend. We calculated the expected mortality rate (E-AAMR) and its 99% CI for each year as the product of the EMRR (99% CI) for that year multiplied by the AAMR for the year 2000. We calculated the number of expected deaths from the E-AAMR for each year in the resident population.

We used the same procedures to obtain data from the United States for resident population in each year; number of recorded deaths; AAMRs; mortality rates from suicide, and mortality rates for cardiovascular, oncologic, and respiratory disease (all adjusted by age); OMRRs; regression equations; EMRRs; E-AAMRs; and number of expected deaths. We repeated the calculations (resident population in each year, number of recorded deaths, AAMRs, adjusted deaths, and OMRRs) for both sexes in Spain and all its 17 regions plus 2 Spanish autonomous cities of Northern Africa (Ceuta and Melilla).

To analyze the evolution of mortality rates in the United States and Spain, we used Pearson’s correlation coefficient and Cohen’s d as an effect size of the differences. Also, we determined the differences in OMRR trends according to Snedecor’s F statistic. We obtained statistics by using SPSS version 21 (IBM, Somers, NY). Finally, we used OpenEpi version 3.01 (Emory Rollins School of Public Health, Atlanta, GA) to estimate the rate ratios as the relative risks (RRs) of mortality with 99% CIs. For graphics, we truncated the y-axis outside the range of values reached by any mortality rate. In addition, because the Spanish National Statistics Institute states that in 2011 they changed the standard population to calculate the Spanish AAMR, we also obtained the trend graphic for the Spanish crude mortality rates.

RESULTS

The overall mortality rate in Spain in 2000 was 888.9 per 10⁵ inhabitants, close to the figure in the United States for that year (854.0/10⁵ inhabitants). The AAMRs were even more similar, at 878.0 per 10⁵ in Spain and 869.0 per 10⁵ in the United States, and diminished in both countries from 2000 to 2010 (r = 0.98; P < .001; Figure 1a; Cohen’s d = −0.228). The OMRR decreased in Spain during this period, reaching its lowest value in 2010 (RR 700/878 = 0.7973; 99% CI = 0.7923, 0.8023; Table 1); and it decreased, too, in the United States (RR 747/869 = 0.8596 in 2010; 99% CI = 0.8576, 0.8616; Table 2).

FIGURE 1—

Evolution of Age-Adjusted Mortality Rates (AAMRs) for (a) Overall, (b) Cardiovascular Disease, (c) Cancer, (d) Respiratory Disease, and (e) Suicide: United States and Spain, 2000–2015

Note. We truncated the y-axis outside the range of values reached by any mortality rate.

TABLE 1—

Evolution of the Population and Mortality Trends in Spain From 2000–2015

Year	Resident Population, No.	Recorded Deaths, No.	AAMRa	Adjusted Deaths,b No.	OMRRc	EMRRd	E-AAMRe	Expected Deaths, No.
2000	40 544 387	360 391	878.0	355 980	1.00	1.00	878.0	355 980
2001	40 766 049	360 131	853.0	347 734	0.97	0.97	853.0	347 734
2002	41 423 520	368 618	849.0	351 686	0.97	0.97	849.0	351 686
2003	42 196 231	384 828	861.0	363 310	0.98	0.98	861.0	363 310
2004	42 859 172	371 934	808.0	346 302	0.92	0.92	808.0	346 302
2005	43 662 613	387 355	816.0	356 287	0.93	0.93	816.0	356 287
2006	44 360 521	371 478	761.0	337 584	0.87	0.87	761.0	337 584
2007	45 236 004	385 361	766.0	346 508	0.87	0.87	766.0	346 508
2008	45 983 169	386 324	747.0	343 494	0.85	0.85	747.0	343 494
2009	46 367 550	384 933	724.0	335 701	0.82	0.82	720.0	337 085
2010	46 562 483	382 047	700.0	325 937	0.80	0.80	702.4	327 055
2011	46 736 257	387 911	893.0	417 355	1.02	0.79	691.9	323 351
2012	46 766 403	402 950	904.8	423 142	1.03	0.77	674.3	315 348
2013	46 593 236	390 419	851.9	396 928	0.97	0.75	656.7	305 998
2014	46 455 123	395 830	841.8	391 059	0.96	0.73	639.2	296 934
2015	46 410 149	422 568	877.4	407 203	1.00	0.71	621.6	288 497

Note. AAMR = age-adjusted mortality rate; E-AAMR = expected age-adjusted mortality rate; EMRR = expected mortality rate ratio; OMRR = observed mortality rate ratio. Rates are given per 100 000 inhabitants.

^aAge-adjusted overall mortality rates.

^bDeaths estimated according to AAMR.

^cAAMR_year/AAMR₂₀₀₀.

^dAAMR_year/AAMR₂₀₀₀, projected on the basis of values before 2010.

^eObtained from the EMRR.

TABLE 2—

Evolution of the Population and Mortality Trends in the United States From 2000–2015

United States
Year	Resident Population, No.	Recorded Deaths, No.	AAMRa	Adjusted Deaths,b No.	OMRRc	EMRRd	E-AAMRe	Expected Deaths, No.	Mortality Rate Ratio Spain/United States, RR (99% CI)
2000	281 421 906	2 403 351	869.0	2 445 556	1.00	1.00	869.0	2 445 556	1.010 (1.006, 1.015)
2001	284 968 955	2 416 425	858.8	2 447 313	0.99	0.99	860.3	2 451 616	0.993 (0.989, 0.998)
2002	287 625 193	2 443 387	855.9	2 461 784	0.98	0.98	851.6	2 449 474	0.992 (0.987, 0.997)
2003	290 107 933	2 448 288	843.5	2 447 060	0.97	0.96	834.2	2 420 196	1.021 (1.016, 1.025)
2004	292 805 298	2 397 615	813.7	2 382 557	0.94	0.95	825.6	2 417 254	0.993 (0.988, 0.998)
2005	295 516 599	2 448 017	815.0	2 408 460	0.94	0.93	808.2	2 388 276	1.001 (0.997, 1.006)
2006	298 379 912	2 426 264	791.8	2 362 572	0.91	0.92	799.5	2 385 488	0.961 (0.957, 0.968)
2007	301 231 207	2 423 712	775.3	2 335 446	0.89	0.90	782.1	2 355 929	0.988 (0.983, 0.993)
2008	304 093 966	2 471 984	774.9	2 356 424	0.89	0.89	773.4	2 351 893	0.964 (0.960, 0.969)
2009	306 771 529	2 437 163	749.6	2 299 559	0.86	0.87	756.0	2 319 285	0.966 (0.961, 0.971)
2010	308 745 538	2 468 435	747.0	2 306 329	0.86	0.86	747.3	2 307 379	0.937 (0.933, 0.942)
2011	311 591 917	2 515 458	741.3	2 309 831	0.85	0.84	730.0	2 274 496	1.205 (1.199, 1.210)
2012	313 914 040	2 543 279	732.8	2 300 362	0.84	0.83	721.3	2 264 168	1.235 (1.229, 1.240)
2013	316 128 839	2 596 993	731.9	2 313 747	0.84	0.81	703.9	2 225 199	1.164 (1.159, 1.169)
2014	318 857 056	2 626 418	724.6	2 310 438	0.83	0.80	695.2	2 216 694	1.162 (1.157, 1.167)
2015	321 418 820	2 712 630	733.1	2 356 321	0.84	0.78	677.8	2 178 641	1.197 (1.192, 1.202)

Note. AAMR = age-adjusted mortality rate; CI = confidence interval; E-AAMR = expected age-adjusted mortality rate; EMRR = expected mortality rate ratio; OMRR = observed mortality rate ratio; RR = relative risk. Rates are given per 100 000 inhabitants.

^aAge-adjusted overall mortality rates.

^bDeaths estimated according to AAMR.

^cAAMRyear/AAMR₂₀₀₀.

^dAAMR_year/AAMR₂₀₀₀, projected on the basis of values before 2010.

^eObtained from the EMRR.

From 2000 to 2010, mortality from cardiovascular (r = 0.99; P < .001; Figure 1b), oncologic (r = 0.98; P < .001; Figure 1c), and respiratory diseases (r = 0.85; P = .002; Figure 1d) decreased in both Spain and the United States (Table A, available as a supplement to the online version of this article at http://www.ajph.org). By contrast, mortality from suicide diverged during this period (r = −0.68; P = .030; Figure 1e; Table A); however, during the second decade of the current century, the mortality in Spain showed a sharp increase beginning in 2011, and the decline in mortality from suicide was also reversed in 2012 (Figures 1a–1e). During this period, mortality in the United States continued to show the declines recorded in the preceding decade for cardiovascular, oncologic, and respiratory disease mortality, whereas mortality from suicide continued to increase (Table 2; Figures 1a–1e). As the mortality statistics in the 2 countries diverged, the significant association of mortality in the correlation analysis disappeared (P = .182) while the effect size of the difference increased greatly after exposure (Cohen’s d = 7.531).

The equation that best explained the trends in the evolution of OMRR in Spain was

(1) OMRR = 1.028 (99% CI = 0.993, 1.064) − 0.020 (99% CI = −0.025, −0.015) × year.

The corresponding equation for the United States was

(2) OMRR = 1.021 (99% CI = 1.005, 1.038) – 0.015 (99% CI = −0.018, −0.013) × year.

Figures 2a and 2b illustrate the evolution of E-AAMRs and their 99% CIs based on these equations, compared with the observed AAMRs in each country. From 2000 to 2010, the difference in trends in OMRRs between Spain (–0.020 by year) and the United States (–0.015 by year) was significant (P = .004); in fact, mortality in Spain was lower than that in the United States during most of the first decade, reaching its lowest value in 2010 with an RR of 0.937 (99% CI = 0.933, 0.942; Table 2). On the contrary, there were not significant differences in these trends between Spain (4.94 × 10⁻⁴ by year) and the United States (4.19 × 10⁻⁴ by year) during the period from 2011 to 2015 (P = .413); however, the sharp increase of mortality in Spain in 2011 has kept this country above the United States during the second decade, reaching its highest value in 2012 with an RR of 1.235 (99% CI = 1.229, 1.240; Table 2).

FIGURE 2—

Evolution of Observed (O) and Expected (E) Age-Adjusted Mortality Rates (AAMRs) in (a) Spain and (b) United States: 2000–2015

Note. CI = confidence interval. We truncated the y-axis outside the range of values reached by any mortality rate.

In the 5-year period from 2011 to 2015, there were 505 559 more deaths in Spain according to the observed AAMR than the expected number of deaths based on the trend from the preceding decade (Table 1). In the United States, the observed number of deaths was only 431 501 greater than the expected number, despite the United States having a population 7-fold as large as Spain (Table 2). For this 5-year period, the observed annual AAMRs in Spain surpassed the expected rates from 2011 (RR = 1.291; 99% CI = 1.283, 1.299) to 2015 (RR = 1.411; 99% CI = 1.403, 1.420); this surpass occurred in both sexes (Tables B and C, available as supplements to the online version of this article at http://www.ajph.org), although it was weaker in women (RR = 1.076; 99% CI = 1.066, 1.086) in 2011 than in men (RR = 1.558; 99% CI = 1.545, 1.57). In the United States, during the same period, the maximum rate ratio between the observed and the expected rates was reached in 2015 (RR = 1.082; 99% CI = 1.079, 1.084).

The sudden increase in AAMR also occurred in all the regions of Spain (Tables D–V, available as supplements to the online version of this article at http://www.ajph.org), in both rich regions (Basque Country: RR 2011/2010 = 1.299; 99% CI = 1.137, 1.484) and poor regions (Extremadura: RR 2011/2010 = 1.222; 99% CI = 1.077, 1.386), and both rural regions (Castile La Mancha: RR 2011/2010 = 1.312; 99% CI = 1.148, 1.499) and urban regions (Madrid: RR 2011/2010 = 1.271; 99% CI = 1.113, 1.45). Finally, the graphic for the Spanish crude mortality rates (Figure A, available as a supplement to the online version of this article at http://www.ajph.org) showed a similar trend to the AAMR.

DISCUSSION

We found a very high figure for excess mortality in the Spanish general population between 2011 and 2015: more than half a million deaths. Mortality statistics for the year 2016 in Spain were not available at the time of writing, but it is reasonably safe to predict that the adjusted number of deaths for this year will be much larger than the expected number on the basis of projections. For the 6-year period from 2011 to 2016, excess deaths attributable to austerity policies are thus expected to surpass the 540 000 deaths attributed to the Spanish Civil War in the first half of the 20th century.¹⁷

The analysis of the increase in the mortality of middle-aged White non-Hispanic population in the United States led to some scientific controversy because it did not include sex- or age-adjusted rates.^18,19 But this is not the case in this study; we have used the age-adjusted rates as provided by governmental institutions in Spain and the United States, and we have shown that the excess mortality affected age-adjusted rates in both sexes and in all regions of Spain. Furthermore, as the Spanish National Statistics Institute changed in 2011 the standard population to compute the AAMR, we have shown that the Spanish crude mortality rates followed a similar trend to the standardized rates. In the end, absolute number deaths overcome any discrepancy on the standard population as we show that the breaking of the mortality trend was of such magnitude that in 2015 Spain suffered 40 000 more deaths than in 2010 despite a decrease in the Spanish population. Currently, the cardiovascular, oncologic, and respiratory diseases cause approximately 70% of deaths in industrialized countries, and all 3 mortality rates increased markedly in Spain after 2010—a finding that suggests that the increase in overall mortality was not a result of a specific group of diseases.

Crises and economic recessions impoverish large sectors of the population and generate health risks. But our results indicate that the effect of financial crisis itself on mortality was not as important as the policies implemented to cope with it. In the United States (and in Spain, but only in 2009 and the first half of 2010), they attenuated the repercussions on mortality, whereas the austerity policies used in Spain since the second half of 2010 exacerbated these consequences. Austerity policies consisted of reduction of government budget deficits through a combination of spending cuts and tax increases, but reductions in government spending during economic downturns increase unemployment, worsen the recession, and expand its length.^20,21 Political mismanagement of financial crises ignores their disastrous effects on human health and has even exacerbated them by adopting harsh austerity measures and cutting key social programs at a time when citizens need them most.²²

When researchers do not distinguish between the effects of the financial crisis and those of the austerity policies, they can state that mortality decreased more during the crisis than before²³; but this statement is misleading from a conceptual perspective.²⁴ Overall mortality in the United States continued to decline despite the economic crisis, and the only change in that country was a slight slow-down in the projected trend, ranging from 2% to 8% of the expected rates. By contrast, after austerity policies were introduced in Spain, a brief induction period sufficed to revert the downward trend seen during the preceding decade such that, since 2011, the effect size of the rates’ differences has been very important and the expected mortality has surpassed the projected rates by 30% to 40%. The cardiovascular,²⁵ oncologic,²⁶ and respiratory²⁵ mortality, as well as suicide,²⁷ are sensitive to changes in health care systems and our results indicate that a very long period is not necessary to detect the effect. This sharp increase in mortality coincides with the increase in socioeconomic inequalities described in the European Union countries that implemented austerity measures in 2010.²⁸

The interaction between fiscal austerity and weakening social protection is what ultimately transforms economic crises into health and social crises.⁸ Spain has faced its first recession in modern history without a national currency, and with budgetary restrictions imposed by the European Union as an external authority. Unable to issue currency or to devalue the euro, in 2011, the Spanish government agreed to amend the Constitution to introduce the “principle of fiscal stability” and to give “absolute priority” to payment of the national debt.²⁹ There are theoretical and historical accounts of the utter failure of austerity programs for dealing with fiscal crises.³⁰ The policy of quantitative easing ran contrary to the fiscal stability recommendations. While the United States’ expansive policies increased the national health expenditure by 22% between 2010 and 2015,³¹ austerity policies in Spain imposed a decrease in the government health expenditure of 7% in the same period.³²

These strict fiscal austerity policies prolonged the economic recession, and the Spanish national health system has become stretched beyond capacity because of budget cuts that have undermined the extent of social protection it is able to provide. The direct effect budget cuts have had on the national health system can be quantified: between 2011 and 2015, 3000 physician jobs, 5000 nursing jobs, and 40 000 other health care positions were lost because of defunding.³³ Until these policies were implemented, the Spanish national health system performed well, according to indicators such as the waiting time for surgery, which was previously similar to that of Sweden³⁴ but which increased after budget cuts were put in effect, as did the waiting time of outpatient consultations of medical specialties.³⁵ Extensive waiting periods contribute to higher mortality and morbidity rates.^36,37

In addition, the United States’ expansive policies introduced the Patient Protection and Affordable Care Act, which, from 2010 to 2015, halved the proportion of uninsured Blacks and Hispanics.²⁶ On the contrary, under austerity policies in Spain, the Royal Decree-Law 16/2012 excluded from public health care large population groups, such as unauthorized residents.³⁸ In short, the Spanish welfare state was weakened at precisely the time it was needed to assist workers negatively affected by unemployment and poverty.

In parallel with this deterioration in the health care system, the unemployment rate for people aged 16 to 64 years in Spain reached 28% at the end of 2012, and at the end of 2016 was still higher than at the end of 2009.³⁹ Under the expansive policies in the United States, the employment cost index increased by 11% during the 2010-to-2015 period,⁴⁰ while it decreased by 0.4% in Spain under the austerity policies.⁴¹ The causality chains from austerity policies during economic crises to the health repercussions are complex; however, these policies have impoverished millions of people, and poverty generates diseases because it imposes poor diets, undernourishment, lack of hygiene, overcrowding at home, and difficulty in accessing health services and education, and impedes a healthy lifestyle.

After 2014, no further austerity measures have been introduced in Spain, possibly because several general and regional elections have been held since then—none of which gave any of the current political parties a parliamentary majority large enough to impose new austerity measures. The economy has begun to recover, but unemployment remains high at 18%³⁹ and workers in Spain are now poorer than before the crisis.¹¹ The relationship between unemployment and suicide is known^15,16 and, as expected, the overall decreasing trend of suicide in Spain up until 2011 was replaced by a sharp upturn between 2012 and 2015; this increase happened in both sexes,⁴² although in some regions, such as Madrid, it did not occur until 2013.⁴³

The causal relationships between mortality and purported risk factors are not easy to establish for small annual variations in mortality rates in a given country, such as those reported in the United States for the year 2015; we still do not know if new phenomena such as the opioid crisis in the United States,⁴⁴ which has not yet occurred in Spain, had some influence on this slight increase, but it is also likely that the expansionary policies did not benefit all the US population. However, large changes in mortality figures such as those seen in Spain after 2010 offer clearer clues as to their causes, given that they are likely to be explainable only by major disasters such as war,¹⁷ breakdown of the state,⁴⁵ or—as in early 21st-century Spain—fiscal austerity policies, employment precariousness, and social vulnerability. Recent European history is replete with examples of how national policies have led to large differences between countries in health and life expectancy.⁴⁶ The gains that can be achieved with social protection policies have also been seen in the United States, where the Social Security Program created in 1935 reversed historical trends in life expectancy for people aged older than 65 years, leading to notable increases.⁴⁷

Limitations

One potential limitation of our study is that national agencies in Spain¹³ and the United States¹⁴ base their mortality statistics on figures that are standardized to the European or US populations. However, the use of different reference populations would not be expected a priori to affect the outcomes we investigated here, in as much as we did not set out to determine which of the 2 countries had a higher mortality rate, but rather to compare the evolution of mortality before and after the implementation of divergent policies in each country. Another potential limitation is our use of linear trends models to project the expected mortality rates, but the linear models showed the lowest residuals among a wide range of trends models that we adjusted for the 2000-to-2010 period. In addition, as supporting information, we offer the monthly mortality rates database (Appendix A, available as a supplement to the online version of this article at http://www.ajph.org) so that this linear trend can be corroborated with more data for that period. We believe that the assumptions that underlie our choice are valid ones, in light of the historical concept of the number of deaths observed above the expected number as described by William Farr.⁴⁸ Over the very long term, mortality rates change slowly, and their secular trends show more marked changes only when causes such as those we analyzed here are brought to bear. We also recognize as a limitation that stronger conclusions could be derived from a study that included a larger number of countries; however, we have compared 2 very large populations that were subjected to absolutely opposite policies, and the introduction of more countries would make necessary the analysis of many more confounding factors according to the different degrees of austerity and expansion in each country.

Conclusions

We conclude that the period from 2011 to 2015 in Spain was characterized by marked excess mortality. There is strong evidence here that points in the direction of the highly adverse public health consequences of austerity policies; the focus on just 2 countries with strongly divergent policy responses helpfully “crystallizes” the matter, but further studies to broaden the evidence base are clearly needed. In the meantime, in light of all the available evidence, governments pursuing austerity agendas will do so in the face of increasingly strong evidence of the negative health impacts on the populations they claim to represent and to serve, including substantial excess mortality.