A Germany without fiscal transfers
Germany shifts an enormous amount of fiscal transfers across jurisdictions each year. This column argues that limits the amount of economic disparities across regions, but comes at the price of lower national productivity and output. Still, with regards to welfare, Germany wouldn’t normally be better off if all fiscal transfers were abolished.
Many countries conduct policies that shift tax revenue across regions so as to tackle spatial economic disparities of their territory (e.g. Acconcia et al.2015). Fiscal transfers are particularly pronounced in Germany. Using data on generated taxes and available public funds at the neighborhood level, in a fresh paper we estimate a substantial overall amount of €65.7 billion was shifted over the 411 districts (Landkreise) in the entire year 2010 alone (Henkel et al.2018). That is equal to 12.4% of aggregate tax revenue and includes various grants from upper- to lower-level government layers, the horizontal fiscal equalisation scheme across Federal States (Länderfinanzausgleich), and extra lower-tier schemes that redistribute from rich to poor jurisdictions within States. The quantity of fiscal transfers within Germany is, thus, a lot more than doubly large as all EU structural funds combined.
Figure 1 Net fiscal transfers across German jurisdictions, 2010
The direction of these fiscal flows is illustrated in Figure 1. The primary recipients are poor and remote locations, often in East Germany, where in fact the inflows take into account up to 23% of local GDP occasionally, or €3,800 per inhabitant. The major net donors are large cities in West Germany with high productivity and wages. Frankfurt contributes 13.3% of its local GDP, or €11,000 per inhabitant, accompanied by Düsseldorf with €6,800. Berlin is a notable exception, since it is a net transfer recipient (of roughly 5% of local GDP) despite being Germany’s largest city.
The impact of fiscal transfers on the spatial economy
With all this strong prevalence of fiscal transfers, surprisingly little is well known about their economic implications. The transfers obviously allow recipient areas to provide more public goods than they otherwise could, as the donors have fewer resources left to invest. The policy may thus deliver what it intends to attain, namely, to lessen intranational economic disparities. However, the question is whether this calls for an equity-efficiency trade-off, namely, that the bigger amount of cohesion comes at the expense of lower national output overall. Ultimately, we want in the aggregate welfare ramifications of this policy, i.e. whether Germany all together would benefit if all fiscal transfers were powered down.
From a purely theoretical viewpoint, there are no obvious answers to those questions. On the main one hand, by shifting resources from rich to poor places, transfers may distort incentives and induce some workers or firms to find in regions that they otherwise wouldn’t normally have chosen (Kline and Moretti 2014). These misallocations could be particularly severe when recipient areas have low degrees of productivity, or if they are remotely located in order that transport losses in interregional trade are exacerbated. The fiscal transfers may therefore reduce aggregate national income, because they divert economic activity from core cities (Hsieh and Moretti 2017). Alternatively, the spatial economy may be suffering from various externalities that folks do not consider when coming up with location decisions. For example, individuals may ignore their effect on others that’s transmitted via local price index effects or through different agglomeration and congestion forces. A laissez-faire equilibrium without transfers may therefore be characterised by an inefficient spatial structure, and specifically, by cities that are ‘too large’ from a social viewpoint (Henderson 1974, Albouy et al.2017). By reducing over-congestion in cities, the fiscal transfers may therefore actually mitigate instead of exacerbate misallocations.
To reveal those important trade-offs, we create and quantify an over-all equilibrium model with multiple asymmetric regions, costly interregional trade, and imperfect labour mobility. More specifically, we build on the recent framework by Allen and Arkolakis (2014) and extend this process to add inter-jurisdictional fiscal transfers. We calibrate the model for Germany, considering taxes and net transfers across regions as seen in (or recovered from) the info. The initial situation inside our model represents a spatial equilibrium, in the sense that utility is equalised across space and people, who are permitted to have idiosyncratic locational preferences, and also have no incentives to go anymore. In a counterfactual analysis, we then simulate how this equilibrium would change if Germany were to abandon all fiscal transfers completely. In this hypothetical scenario, local public goods are financed solely by taxing local economic activity, but there are forget about transfers across jurisdictions. Comparing this counterfactual to the actual equilibrium we can provide quantitative answers to the questions of how, and by which channels, fiscal transfers affect the spatial economy.
Abolishing transfers raises national output however, not welfare
Inside our counterfactual analysis, we first observe a significant migration wave out of your former recipient and towards the former donor regions. Some East German cities would lose almost one quarter of their population, while big cities like Frankfurt or Munich would grow substantially. Altogether, we calculate that roughly 3.2 million people would change their district of residence. This number of cross-district migrants is a lot more than thirty times bigger than the actual migration that people typically observe each year. Fiscal transfers therefore appear to substantially affect people’s location choices.
As the induced migration from abolishing transfers is from less to more productive regions, we observe substantial gains in aggregate output and productivity at the national level – inside our benchmark specification, we calculate an increase in real GDP per capita of 3.7%. Average labour productivity would increase by 5.8%. The most notable panel in Figure 2 illustrates the counterfactual change in national productivity for various parameter constellations. It really is consistently positive and will range up to almost 20%. Fiscal transfers thus appear to imply an equity-efficiency trade-off – by retaining economic activity in the periphery, they limit productivity and income dispersion across space. But this comes at the expense of lower national productivity and output.
Figure 2 Impact of abolishing fiscal transfers on national productivity and welfare
When embracing national welfare, however, we find that trade-off could possibly notexist. Inside our baseline counterfactual, welfare even mildly decreases by 0.05% when transfers are abolished. In other parameter constellations, illustrated in underneath panel of Figure 2, we sometimes obtain welfare losses and sometimes gains, based on parameters. But even though there are welfare gains, they are consistently an order of magnitude smaller compared to the increases in national productivity and output. Quite simply, abolishing fiscal transfers from rich to poor regions within Germany increase national GDP, nonetheless it may decrease welfare.
What’s the intuition because of this surprising difference? The effect is rooted in inefficiencies of the original spatial equilibrium. One important channel inside our model is that the former donor regions already are ‘too large’ from a social perspective. There is a mixture of negative and positive externalities inside our framework that folks ignore within their private location decisions, representing different agglomeration and congestion forces as well as the sharing of local public goods. However the combined net externality that single individuals impose on others should be negative at the margin, otherwise it really is difficult to rationalise why the economic geography of a country will not collapse right into a single city. Hence, large cities have a tendency to be ‘too large’, although interregional transport costs and market size effects could work from this tendency.
The fiscal transfers that people observe the truth is effectively countervail over-congestion, because they offer incentives for workers to reside in beyond your big cities. This could be welfare-enhancing, particularly if transport charges for goods are not too big and if the recipient areas aren’t too remotely located. And vice versa, abolishing the prevailing transfers can be harmful to welfare, regardless of the associated productivity gains, since it makes the issue of over-congestion in large cities a whole lot worse. The fiscal transfers are surely, for the most part, a second-best instrument and don’t implement an ‘optimal spatial structure’ of the economy in virtually any meaningful sense. But our results claim that they could shift the economy nearer to this optimum.
One might argue that policymakers should first counteract the tendency of over-congestion, for instance by detatching housing supply restrictions in popular cities or through investments in urban transport systems, and reconsider the case for redistributive fiscal transfers. But such a proposal misses the idea. You can easily suppose urban infrastructure or construction investments are advantageous to society and easily pass standard cost-benefit assessments. But individuals will respond to those changes. The spatial equilibrium calls for more folks moving to the large cities. However, this will not change the logic that the web agglomeration externality is negative at the margin even in the brand new equilibrium. The case for transfers may thus remain unchanged even following the urban investments took place.
Our results could have important implications for spatial redistribution policies in Germany and elsewhere, specifically because they claim that equity-efficiency trade-offs can be extremely different with respect to the economic variable (output versus welfare) one is concentrating on.
Albouy, D, K Behrens, F Robert-Nicoud, and N Seegert (2017), “The perfect distribution of population across cities”, unpublished manuscript, University of Illinois.
Allen, T, and C Arkolakis (2014), “Trade and the topography of the spatial economy”, Quarterly Journal of Economics 129(3): 1085-1139.
Henderson, J V (1974),“The sizes and types of cities”, American Economic Review 64(4): 640-656.
Henkel, M, T Seidel, and J Suedekum (2018), “Fiscal transfers in the spatial economy”, CEPR Discussion Paper 12875.
Hsieh, C-T, and E Moretti (2017), “Housing constraints and spatial misallocation”, unpublished manuscript, University of California, Berkeley.
Kline, P, and E Moretti (2014), “Local economic development, agglomeration economies and the big push: a century of evidence from the Tennessee Valley Authority”, Quarterly Journal of Economics 129(1): 275-331.