Social and economic multipliers: What they are and why they are important for health policy in Europe

Introduction

Public health has, at its heart, the imperative of preventing disease. Acting early, in those circumstances where it is possible to prevent ill health, will always be better than trying to treat an established disease, both in terms of the human suffering avoided but also, in many cases, the lower ultimate cost.

Yet not everyone agrees. Consider financial decisions being made across Europe. In the pursuit of ‘austerity’, a programme to reduce deficits and debts, governments in many parts of Europe are making deep budget cuts that are reducing public health measures that are known to be effective. These fiduciary decisions are often made with little regard to their health consequences or even their longer-term social and economic ramifications.

In a time of budget pressure, it is more important than ever that those working in public health are able to engage with finance ministers, not only using health arguments but also economic ones.

Here, we review some of the conventional methods used to make the economic case for investing in health and then go beyond them to argue for a ‘multiplier approach’ to health investment, which is more in line with macroeconomic perspectives.

Conventional approaches: ‘Cost-of-illness’ studies

Most decision-makers have seen commonly cited figures about the costs of obesity, diabetes, cancer and other common chronic diseases. For example, the UK has assessed the cost of heart disease to the NHS and UK economy at £30b (~€50b) per annum [1]. The methods used to produce these figures, termed cost-of-illness studies, are usually simple and straightforward. Indeed, this is their major advantage.

The approach taken is as follows [2]. First, researchers will add up how much it costs to treat heart disease, based on the medical care, drugs, services, and technologies delivered. These medical care costs are usually referred to as the ‘direct costs’ of illness to the health system. Then, if they wish to go further, the researchers will investigate the wider costs to society. These are known as the ‘indirect costs’ of illness. They may calculate the economic cost of missing work days due to heart disease, ‘absenteeism’, or of early retirement; or they may go even further to calculate the loss of productivity at work due to the illness, ‘presenteeism’. Finally, these sums are added – depending on the approach taken – to give the overall estimate of the cost-of-illness for a specific disease.

Cost-of-illness approaches often generate large and headline-grabbing figures. They can help draw attention to the magnitude of the problem. But they do not fully portray the true financial burden of disease, and may overestimate the problem.

Cost-of-illness approaches generally have difficulty addressing important policy questions: Could these healthcare and wider economic costs be avoided? If, for example, people avoid dying suddenly from heart disease, but instead get cancer later, have we saved money in the long-term?

Cost-of-illness studies also produce results that do not fit with budget sheets of finance ministers. Those responsible for national budgets will want to know, for example, if investing in heart disease prevention will help them reduce their budget deficit or make it worse. The prevailing approach of cost-of-illness does not tell us how much government spending will go up or down in a given year because it does not necessarily consider the relative effectiveness of different interventions. So, as far as a budget planner is concerned, asking for an additional €10m for heart disease prevention might be a good way to improve health and may even save money, in the long-term; but the timing of those savings may require a short-term increase in the annual budget deficit. The guaranteed short-term deficit coupled with the uncertainty of the long-term savings make this a difficult sell.

To engage more effectively, we need to start from the decisions that finance ministers are grappling with. This is a shift in our usual thinking. It involves addressing questions such as, ‘will investing in public health help reduce unemployment rates?’, ‘will investing in public health reduce government spending, and, if so, when?’, ‘what impact will public health spending have on economic growth?’, and ‘how does public health spending compare with other areas of government spending, such as investment in defence, environment, and other areas of social expenditure?’

These questions, at present, lack good answers, not just in data-poor regions of Europe (such as the South East) but across the data-rich, advanced industrialised nations as well.

The multiplier approach

Before going into the details of a macroeconomic approach, it is helpful to review the background to perhaps the biggest economic debate of our time: austerity versus stimulus.

Governments face tough decisions in two areas of macroeconomic policy: whether and how much money to spend, and whether to increase the amount of money in circulation. These two tools, fiscal and monetary policy, respectively, are the main pillars of government economic policy. Here we’ll focus mainly on fiscal policy, which seeks to control how much the government spends and on what.

First, however, we need to consider the backdrop. In the aftermath of the financial crisis of 2008, European governments spent a colossal amount bailing out the banks. In the UK this totalled over £1 trillion. The result was that governments took private debt from the banks, and absorbed it onto their balance sheets as public debt. The cost of servicing this debt then increased annual budget deficits. The problem was compounded by slowing economies, so that less tax revenue came in to the government from taxes, and especially from the UK financial sector that avoided paying taxes by declaring losses. Declining revenues coupled with rising spending – due to greater spending on social security, e.g. unemployment benefit – further increased the budget deficit.

Several options were available. To reduce the deficit, governments could try to boost the economy, so that more revenue would be available because of increased tax revenues. Or it could increase taxes, of various types, again to generate more revenue. Or it could spend less on its services and people.

Across Europe a narrative was created that governments had spent too much money and needed to cut back; a narrative that was cheered on by sections of the media. For example, in the UK, 55% of the population blamed excessive spending by the government in power prior to the crisis for current government debt, while only 45% blamed the decision to bail out the banks (authors’ analysis: British Social Attitudes Survey 2016 edition). Against this background, it was unsurprising that policy focused on budget cuts as a means to solve the problem.

Intuitively, we might expect that reducing government spending would automatically lower the deficit. However the calculus is not so simple because this assumes that tax revenues would remain stable. But government spending affects many economic forces, including consumer spending and consumption, business sales and employment, and choices by firms on whether to invest money in one country or another, among others. All of these also influence tax revenues. Thus, if government cuts spending by €100 but revenues fall by €110 then the deficit remains and the debt continues to rise.

Hence, a simple approach such as that used in ‘cost-of-illness’ studies similarly cannot answer budgetary questions on whether to make cuts or go for a stimulus. We need to know more about how government spending affects the economy.

Enter the ‘fiscal multiplier’. The entire debate about austerity – whether to cut budgets or to spend more – hinged on whether or not government spending would increase or decrease economic growth. The fiscal multiplier statistic describes the impact of each euro of government spending on economic growth. It is the velocity of money in the economy; as each euro moves from your pocket to a supermarket to another employee to a butcher to another firm and so on. The faster a single pound moves around the economy in a given time period, the more economic growth it creates. When the multiplier is greater than 1, it means that each additional euro spent translates into more than 1 euro in economic growth. For example, if the multiplier is 2 then spending an additional euro will increase GDP by 2, thereby increasing tax revenues. When the multiplier is less than 1 (but greater than zero), then the economy still gets larger but each additional euro spent fails to multiply across the economy, delivering slower growth.

Initially, international financial institutions, such as the International Monetary Fund (IMF), based their advice on the assumption that the fiscal multiplier was about 0.7. This meant that austerity would be good for the economy because, currently, spending each 1 euro was only returning 0.70 euro cents – leading to slower growth, smaller increases in tax revenues, and greater deficits. On this basis, the IMF, European Commission, and European Central Bank (the ‘troika’) recommended that Greece, among others, make deep cuts, projecting that a 4% budget cut could yield a return of as high as 2−3% in economic growth [3].

Soon the results of this austerity experiment became apparent. Rather than grow, Greece’s economy continued to shrink. An intense debate broke out as to why but it soon became clear that there was a problem with the IMF’s assumptions [4]. In late 2013 it re-estimated the fiscal multiplier using data from the crisis itself. This analysis demonstrated very clearly that the multiplier was much greater, between 1.2 and 1.7 − suggesting greater public spending would have helped not harmed the recovery [3,5].

Yet while these new figures are good for increased spending, what should finance ministers be spending on? Governments needed to make informed, data-driven choices about managing their economies. To inform this debate, our team compiled macroeconomic data from across Europe and re-estimated the multiplier once again, but this time disaggregating the government expenditure data into its components, such as health, social protection, education, defence, and so forth [5].

This new analysis produced the same overall figure as that estimated by the IMF, an overall spending multiplier in Europe of about 1.7 – so each 1 euro spent, in the short term, increased economic growth by €1.70. However, more importantly, the size of the multiplier varied widely across government sectors. The multipliers were especially large, up to ~€3, in health, education, and social protection, representing a large, short-term return on investment. In contrast, multipliers associated with defence spending and bank bailouts were smaller and in some cases negative. When we looked more closely we found that, in these cases, money was flowing out of the economy, perhaps to US firms for spending in conflict zones in the case of defence or paying down international debts in the case of bailouts. In contrast, money spent on health and education was flowing into the pockets of often low-paid workers, who in turn were spending much of it in local shops, which were buying from local suppliers and using local services.

This information can guide government’s budgetary decisions, suggesting areas where public expenditure will boost growth and increase tax revenues.

This short example demonstrates the need for the public health community to shift our thinking if we are to engage in economic debates that impact public health.

For health advocacy, the point is that if even a detailed macroeconomic analysis cannot inform a critical macroeconomic debate, how can we expect the cost-of-illness approaches commonly employed in health economics to shift key government decisions about whether or not to invest in better health?

We need to go beyond our traditional approaches and assess the health spending ‘multipliers’ for treatment and prevention. In the next section we outline how this could work.

Social and environmental multipliers

Whether it is a vaccination programme or a hospital construction project, every public health project impacts the wider economy. In other words, there is a multiplier effect.

These multipliers are of course economic – for the health sector is in many countries the biggest employment and economic sector – but they are also social. Health programmes impact people’s educational attainment, the quality of the environment, and their levels of ‘social capital’ (trust in their neighbours, friends, and family), among others.

Example of estimating multipliers: Hospital construction in Novi Pazar, Serbia

How can we assess these multipliers? To start, consider a textbook model for engaging with policymakers and what the multiplier approach could look like, while realising this ideal case may be difficult to reproduce given existing data limitations.

Suppose, as a thought experiment, a health minister wishes to build a new 210-bed hospital in a rural, deprived community of Southern Serbia, such as around Novi Pazar. The cost would be around US$100m for the hospital itself, plus another US$40m for roads, equipment and infrastructure.

The finance minister wants to know: Is this a good investment? Is a hospital really needed? How could the construction and maintenance plans be kept affordable?

The minister is concerned about a wider set of problems in the region too. The south of Serbia has suffered drought and shrinking populations, and is grappling with tourism as a potentially beneficial economic sector, while also needing to protect the environment.

To answer these questions, and in the absence of data, researchers could draw from existing studies, perhaps from other parts of the country, or neighbouring countries, to assess how building a hospital historically had impacted on the local community. A hospital provides jobs for health and other workers, and can also attract investment, as it may make the community a more desirable place to live and work. Indeed, several western European countries have explicitly invested in health and education facilities to compensate for the loss of heavy industry in some locations, such as Maastricht in the Netherlands. It is important to account for all these fiscal and social multipliers.

It would be important for the finance minister to know that, in rural settings, hospitals provide a huge return on investment. A classic study in the US found that rural hospitals contribute, on average, $700,000 to $1m to local economies, as a result of retail sales, tax collection, employment benefits, and wider economic services. This exceeded the operating cost by $54,000 per hospital bed. They estimated that, as a result, ‘from a purely economic standpoint it would make sense to subsidise the hospital’ [6]. We can learn this from times when hospitals have been opened, or, in reverse, when they have been closed. For example, a study of hospital closures in the United States found that closing a sole hospital in a community reduced per-capita income by $703, or 4%, and increased the unemployment rate by 1.6 percentage points [7]. The researchers concluded that ‘the local economic effects of a hospital closure should be considered when regulations that affect hospitals’ financial well-being are designed or changed’. In other words – it’s necessary to estimate the fiscal and social multipliers.

Ideally, methods such as these would be used to inform the decision about opening the hospital in Novi Pazar, in addition to conventional health needs assessments. Moreover, they should be linked to best practice in maximising the local benefits, such as designing contracts in ways that support local suppliers. Such measures have benefits that go beyond the economic, to include the environment, with shorter supply chains.

After these projections of a potential health dividend have been made, and assuming the project goes ahead, then it is important to measure the extent to which it is being realised. Ideally a study would be designed to cover pre- and post- the hospital’s construction to take stock of the impact of the hospital on not just the community’s health but the community’s economic wellbeing. The methods employed are similar to those used in epidemiology for evaluating interventions – only using different economic endpoints and outcomes in our equations.

There are at least four important things to consider when estimating multipliers (see Box).

Conclusion

This review makes the case for shifting approaches to estimating the returns on investment in health from simple cost accounting to multiplier approaches that are in line with techniques used by macroeconomists. Our hope is to improve the ability of public health practitioners to engage with economic planners. If multiplier approaches are employed to evaluate whether or not to host major events such as the Olympics or engage in war or peace, why not apply the same logic to investments in health promotion?