Starting in July 2018 the BAG – Bundesamt für Güterverkehr introduced yet another policy change which affected how lorries pay tolls within the MAUT system as well as the data that come out of this process which are used for computing the Toll Index. The change expanded the network of roads in which toll is due by adding all bundesstraßen to it. While in the long run this is bound to make the Toll Index more accurate this introduced a discontinuity. The BAG even had difficulty producing the numbers timely for about year. After July 2019 we can report year on year changes for each month (with a missing value in 2018 for all months from July to December and a missing value in 2019 for all months from January to June. The Toll Index was first proposed in IZA DP5522 which was published in the Journal of Forecasting. It has been widely covered in national and international media (selection):
The German statistical office, in cooperation with the Bundesamt für Güterverkehr, has taken the MAUT data in its portfolio of data products and their efforts can be found here. The Destatis document describing the data is here and here is their publication calendar for 2019.
Citation: “Nowcasting business cycles using toll data.” Journal of Forecasting 32:4 (2013): 299–306(with K. F. Zimmermann).
Starting in July 2018 the BAG – Bundesamt für Güterverkehr introduced yet another policy change which affected how lorries pay tolls within the MAUT system as well as the data that come out of this process which are used for computing the Toll Index. The change expanded the network of roads in which toll is due by adding all bundesstraßen to it. While in the long run this is bound to make the Toll Index more accurate this introduced a discontinuity. The BAG even had difficulty producing the numbers timely for about year. After July 2019 we can report year on year changes for each month (with a missing value in 2018 for all months from July to December and a missing value in 2019 for all months from January to June. The Toll Index was first proposed in IZA DP5522 which was published in the Journal of Forecasting. It has been widely covered in national and international media (selection):
The German statistical office, in cooperation with the Bundesamt für Güterverkehr, has taken the MAUT data in its portfolio of data products and their efforts can be found here. The Destatis document describing the data is here and here is their publication calendar for 2019.
Citation: “Nowcasting business cycles using toll data.” Journal of Forecasting 32:4 (2013): 299–306(with K. F. Zimmermann).
Starting in July 2018 the BAG – Bundesamt für Güterverkehr introduced yet another policy change which affected how lorries pay tolls within the MAUT system as well as the data that come out of this process which are used for computing the Toll Index. The change expanded the network of roads in which toll is due by adding all bundesstraßen to it. While in the long run this is bound to make the Toll Index more accurate this introduced a discontinuity. The BAG even had difficulty producing the numbers timely for about year. After July 2019 we can report year on year changes for each month (with a missing value in 2018 for all months from July to December and a missing value in 2019 for all months from January to June. The Toll Index was first proposed in IZA DP5522 which was published in the Journal of Forecasting. It has been widely covered in national and international media (selection): Focus Magazin, Tim Harford – The undercover economist, Financial Times, MoneyWeek, WirtschaftsWoche, CNN International, DRS3 Swiss public radio, Deutsche Welle. The German statistical office, in cooperation with the Bundesamt für Güterverkehr, has taken the MAUT data in its portfolio of data products and their efforts can be found here. The Destatis document describing the data is here and here is their publication calendar for 2019.
Citation: “Nowcasting business cycles using toll data.” Journal of Forecasting 32:4 (2013): 299–306(with K. F. Zimmermann).
Starting in July 2018 the BAG – Bundesamt für Güterverkehr introduced yet another policy change which affected how lorries pay tolls within the MAUT system as well as the data that come out of this process which are used for computing the Toll Index. The change expanded the network of roads in which toll is due by adding all bundesstraßen to it. While in the long run this is bound to make the Toll Index more accurate this introduced a discontinuity. The BAG even had difficulty producing the numbers timely for about year. After July 2019 we can report year on year changes for each month (with a missing value in 2018 for all months from July to December and a missing value in 2019 for all months from January to June. The Toll Index was first proposed in IZA DP5522 which was published in the Journal of Forecasting. It has been widely covered in national and international media (selection):
The German statistical office, in cooperation with the Bundesamt für Güterverkehr, has taken the MAUT data in its portfolio of data products and their efforts can be found here. The Destatis document describing the data is here and here is their publication calendar for 2019.
Citation: “Nowcasting business cycles using toll data.” Journal of Forecasting 32:4 (2013): 299–306(with K. F. Zimmermann).
Starting in July 2018 the BAG – Bundesamt für Güterverkehr introduced yet another policy change which affected how lorries pay tolls within the MAUT system as well as the data that come out of this process which are used for computing the Toll Index. The change expanded the network of roads in which toll is due by adding all bundesstraßen to it. While in the long run this is bound to make the Toll Index more accurate this introduced a discontinuity. The BAG even had difficulty producing the numbers timely for about year. After July 2019 we can report year on year changes for each month (with a missing value in 2018 for all months from July to December and a missing value in 2019 for all months from January to June. The Toll Index was first proposed in IZA DP5522 which was published in the Journal of Forecasting. It has been widely covered in national and international media (selection):
The German statistical office, in cooperation with the Bundesamt für Güterverkehr, has taken the MAUT data in its portfolio of data products and their efforts can be found here. The Destatis document describing the data is here and here is their publication calendar for 2019.
Citation: “Nowcasting business cycles using toll data.” Journal of Forecasting 32:4 (2013): 299–306(with K. F. Zimmermann).
In our paper with K. Tatsiramos and B. Ferheyden published recently in Nature’s Scientific Reports we looked at a number of policies (aka non-pharmaceutical interventions or NPIs) and their implementation intensity across 175 countries. These are:
cancelling public events,
imposing restrictions on private gatherings
closing schools
closing workplaces
restrictions on internal movement
restrictions on public transport had
International travel restrictions
stay-at-home requirements
We studied the differential or marginal (additional) effect of each policy on the pandemic and on population behavioral patterns as expressed by Google mobility data, conditional on other concurrent policies. We found that:
The first four policies (events, gatherings, workplaces and schools) all have both statistically and size-wise significant marginal antivirus effect.
The same is true for stay-at-home requirements with a small caveat in terms of statistical significance.
Restrictions on public transport and internal movement affected population patterns but not the epidemic.
International travel controls had only a short lived effect.
Overall we find that time use considerations and the epidemiologically relevant characteristics of the places affected by the NPIs (numerosity, density, behavioral norms, geographic range, frequency, traceability etc) are the factors which need to be taken into account. This explains 1 above.
Also some policies have collateral effects (they affected secondary places in addition to their primary target) which explains 3: once you removed all destinations, moving made little sense so the density and numerosity at public transport hubs dropped to safe levels. In other words when the public transport was shut down it was like dousing a brush fire already deprived of oxygen.
Staying at home is statistically less significant because the measure came when the shit hit the fan and international travel controls is like a levee which failed to prevent the flood: once breached you still need to mend it but mainly you need to mop up the flood.
A couple of final remarks regarding some of the current debates in Germany and elsewhere.
Schools matter epidemiologically. Children are like bees: if they get pollen they will cross-pollinate flowers. Replace pollen with COVID19 and you get the picture. For those who say children don’t get infected: the share of 5-14 year olds in the German population is 7.4% and the 6-14 year olds make up 5.5% of the COVID19 case as of this writing. To make matters worse what makes kids lovable also makes them more likely to engage in contagious behaviors and schools are dense places while school attendance occurs at high frequency. I recall of reading of a mom whose son returned home from school wearing someone else’s mask… The problem is that schools also matter both educationally as well as economically (long term through loss of education, short term by obstructing working parents if closed). If we want to open schools we need to reduce pollen. If we shut down places which are numerous, dense and support contagious behavioral norms (i.e. weddings, soccer matches, concerts etc), if we do as much home office as possible and protect those working onsite and if we have some form of home curfew then we might be able to reduce pollen enough so as to send our bees to school without making things worse. But whether this works or not is an empirical question. We need to apply NPIs and measure effects.
The 15 km restriction in Germany is both unenforceable as well as useless. I drove to the Eiffel 100 km from home and went for a 20 km hike in the snow, being prudent and cautious in parking places. I did not see a soul in the forest. In contrast when I take a walk in the neighborhood I see more people. In other words: it does not matter how far you go. It matters where you go. In other words is your destination dense, populous etc? It is also relevant how you go there (public transport, car, bike etc).
Decisions must be made on the basis of epidemiologically relevant considerations.
Citation:
Askitas, N., Tatsiramos, K. & Verheyden, B. Estimating worldwide effects of non-pharmaceutical interventions on COVID-19 incidence and population mobility patterns using a multiple-event study. Sci Rep11, 1972 (2021). https://doi.org/10.1038/s41598-021-81442-x