Ifo WORKING PAPERS 341 2020 - ifo Institut
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
ifo 341
2020
WORKING November 2020
PAPERS
The Pass-Through of Temporary
VAT Rate Cuts
Evidence from German Retail Prices
Clemens Fuest, Florian Neumeier, Daniel StöhlkerImprint: ifo Working Papers Publisher and distributor: ifo Institute – Leibniz Institute for Economic Research at the University of Munich Poschingerstr. 5, 81679 Munich, Germany Telephone +49(0)89 9224 0, Telefax +49(0)89 985369, email ifo@ifo.de www.ifo.de An electronic version of the paper may be downloaded from the ifo website: www.ifo.de
ifo Working Paper No. 341
The Pass-Through of Temporary VAT Rate Cuts
Evidence from German Retail Prices*
Abstract
On 1 July 2020, value added tax (VAT) rates were reduced in Germany to fight the
economic consequences of the Corona pandemic. The VAT rate reduction is temporary
as rates will return to their previous level on 1 January 2021. We study the effects of
the temporary VAT rate cut on German supermarket retail prices using an extensive
webscrapped data set covering the daily prices of roughly 190,000 products. To identify
the causal price effects, we compare the development of prices in Germany to those
in Austria. Our findings indicate a nearly full pass-through of the VAT rate reduction
on prices. On average, prices in German supermarket retail decreased by 2% after
the implementation of the VAT rate reduction. We also provide evidence that prices
in more competitive product markets decreased to a larger extent.
JEL Code: E31, H22, H25
Keywords: Value added tax, tax incidence, price effects, competition
Clemens Fuest Florian Neumeier
ifo Institute – Leibniz Institute for ifo Institute – Leibniz Institute for
Economic Research Economic Research
at the University of Munich, at the University of Munich,
University of Munich, CESifo University of Munich, CESifo
fuest@ifo.de neumeier@ifo.de
Daniel Stöhlker**
ifo Institute – Leibniz Institute for
Economic Research
at the University of Munich,
University of Munich,
Munich Graduate School
stoehlker@ifo.de
* This work benefited from comments that we have received at the EconPol Workshop on Public
Policy Evaluation. Ludwig Oetker provided excellent research assistance during early stages of the
project.
** Corresponding author: Daniel Stöhlker.Fuest, Neumeier and Stöhlker
1. Introduction
On 3 June 2020, the German government announced an unprecedented stimulus package
in order to combat the economic consequences of the Corona pandemic, worth in total
EUR 130 billion. In addition to cash transfers to households and tax deferrals for firms, the
main element of the program – to the surprise of the general public – was a temporary
reduction of the value added tax (VAT) rates. Limited explicitly to the second half of 2020,
the standard VAT rate was cut from 19% to 16% and the reduced rate, which applies broadly
to ‘basic food and beverages’, was reduced from 7% to 5%. The VAT rates will return to
their previous level on 1 January 2021.1
Temporary VAT cuts are a relatively new instrument in the fiscal policy toolkit. A
widely cited example is the VAT rate cut in the UK in 2008 and 2009. As explained in
Blundell (2009) and Crossley et al. (2009), the objective of this measure is to generate a
short term boost in consumption. This will only work if the tax cut temporarily reduces
prices, creating incentives for consumers to bring forward spending and, therefore, stim-
ulate aggregate demand in times of crisis.
While the pass-through of permanent VAT changes has been studied extensively (see
e.g. Benedek et al., 2019; Carbonnier, 2007; Kosonen, 2015), very little is known about the
price effects of temporary VAT changes, not least because temporary VAT changes have
been rarely used. There is reason to believe that the pass-through could be different for
temporary VAT changes due to various reasons, most importantly because menu-costs for
firms to adjust prices have to be borne twice when VAT rates are changed only temporar-
ily.
In this study we use daily retail prices from German and Austrian supermarkets in or-
der to track the impact of the temporary VAT reduction in Germany on consumer prices
day by day. Austrian supermarket prices serve as a counterfactual, allowing us to identify
the causal price effects of the German VAT rate cut. Austria has adopted a stimulus pack-
age including similar instruments as Germany, and the stimulus package was adopted at
around the same time. However, Austria did not reduce VAT rates in the retail market.2
We collected daily prices for 190,000 products from REWE, a German supermarket chain,
and for 9,000 products from Billa in Austria using a webscraping algorithm. We find a
large and immediate pass-through of the VAT reduction to prices, starting in the week
before the tax reduction. On average, German supermarket prices decreased by roughly
2%, and the effect persists until today.
We also investigate the relationship between pass-through and the degree of com-
petition in product markets, using the number of brands operating in the same product
sub-category as a proxy for the degree of competition. We find that the pass-through is
notably larger in product segments where more brands are competing with each other,
1 The press release (from June 4 2020) can be accessed under the following link: https://www.
bundesfinanzministerium.de/Content/EN/paStandardartikel/Topics/Public-Finances/
Articles/2020-06-04-fiscal-pasckage.html.
2 Austria reduced the VAT only in the hotel industry.
2Price Pass-Through of Temporary VAT Cuts
thus providing evidence for a positive relationship between the VAT pass-through and
competitive pressure.
To the best of our knowledge, there are only two studies that provide empirical ev-
idence on the pass-through of temporary VAT cuts to consumer prices. Crossley et al.
(2014) analyze the price adjustments in response to the temporary VAT cut by 2.5 per-
centage points for 13 months in the United Kingdom during the 2008/2009 financial crisis.
Their results indicate that the tax cut was initially passed on to consumers in the form of
lower retail prices but after a few months, long before the expiration of the tax cut, prices
increased again. Montag et al. (2020) also focus on the recent VAT cut in Germany and
track the impact on fuel prices. The authors compare price trends at German and French
gas stations and find that the pass-through in the case of diesel fuel is around 80% while it
is smaller for gasoline, arguing that drivers of automobiles with diesel engines drive more
and are therefore more price sensitive. However, according to the “consumer price index
(CPI) basket”, based on which the inflation rate in Germany is calculated, expenditures on
fuel at gas stations account for only 2.5% of all consumption purchases of German house-
holds. In contrast, the products we consider in our analysis account for more than 25% of
German households’ total expenditures. A temporary VAT cut on electricity consumption
in Belgium and its subsequent price effects and consumption adjustments are studied in
Hindriks and Serse (2020). The VAT cut in 2014 was intended (and communicated) to be
a permanent change but was revoked by a different government in the year afterwards.
Given that the reform had no strict expiry date at the time of its implementation, the
reform cannot be counted as a temporary VAT cut in the sense of this paper.
Also related to our study is a strand of the literature that studies the price effects
of sales tax holidays, i.e. temporary exemptions of specific consumption goods, such as
clothes or school supplies, from the state sales tax in the United States. Despite being
typically limited to several days only, previous studies have documented an almost full
pass-through of temporary sales tax exemptions to sales prices (Harper et al., 2003) or
even an over-shifting for some commodity goods (Cole, 2009).
Our results also feed into the discussion about the effectiveness of ‘unconventional
fiscal policies’ (Feldstein, 2002) and the extent to which inflation expectations of house-
holds can be manipulated (D‘Acunto et al., 2018; D‘Acunto et al., 2016) – an idea that has
gained traction when the zero lower interest rate bound and high public debt burdens
started limiting the scope of monetary and fiscal policy.
The rest of the paper is structured as follows: Section (2) describes the price data
from Germany and Austria used in our analysis. Graphical evidence on the VAT pass-
through and the results of the event study analysis are shown in Section (3). The impact
of the extent of competition on the pass-through is discussed in Section (4). Section (5)
concludes.
3Fuest, Neumeier and Stöhlker
2. Data
In order to identify the impact of the temporary VAT reduction, we exploit daily price
data from the online shops of REWE in Germany and Billa in Austria, which is also part
of the REWE group. An automatized webscraping algorithm collects the relevant prod-
uct information every day from both online shops, including the product ID, the product
name, the (sub-) category it belongs to and the current retail product price. Online prices
mirror those that apply in ‘physical’ shops. As explained in Rickert et al. (2018), while
the purchasing activities of supermarket chains are handled centrally by the headquarter,
product pricing is done locally, leaving it up to the discretion of each shop what prices
to charge. In some cases, this means that different shops charge different prices for the
same product. In such cases, the online shops show the price interval for the product. We
decided to use the lower bound of that price interval in our analysis.
In the case of REWE, we collected up to 190,000 product prices per day, starting in
September 2019. Data from Billa is available since 4 June 2020, i.e the first day after the
announcement of the VAT reduction in Germany. It covers around 9,500 products per day.
The assortment of products available to customers in the online shops exceeds the set of
products that are available in physical stores which is mostly restricted to food, beverages,
and hygiene products. Online shops, in contrast, also feature a limited number of clothes,
various electronic devices as well as some kitchen and garden utensils.
Table 1: Coverage of Harmonized Product Sets and Share in Consumption Basket
REWE Billa
Basket Share Abs. # Rel. Share Abs. # Rel. Share
Hygiene, Cosmetics & Cloths 5.72 3985 6.80 697 8.66
Garden & Outdoor 0.62 732 1.25 24 0.30
Non-alcoholic Beverages 0.79 3970 6.78 591 7.34
Home- & Hobby Utensils 4.73 23 108 39.44 976 12.13
Coffee, Tea & Cocoa 0.40 1538 2.63 138 1.71
Kitchen Utensils 0.27 6058 10.34 58 0.72
Food 5.68 12 107 20.66 3753 46.63
Fruits & Vegetables 2.00 425 0.73 201 2.50
‘Sweet & Salty’ 0.71 2455 4.19 702 8.72
Wine, Liquor & Tobacco 3.78 2647 4.52 659 8.19
Pet Supplies 0.43 1564 2.67 250 3.11
Total 25.13 58 589 100.00 8049 100.00
Notes: The set of products refers to those that were available on June 4, 2020, which
is our base day. The composition of the average consumption basket in Germany
can be accessed online at: https://www.destatis.de/DE/Themen/Wirtschaft/Preise/
Verbraucherpreisindex/FAQ/anteil-gueter-warenkorb.html.
4Price Pass-Through of Temporary VAT Cuts
Table (1) provides an overview of the harmonized product set after excluding those
product categories for which there is no comparable counter-part in the product set of
the other supermarket chain. The harmonized data comprise around 60 thousand products
from REWE and eight thousand products from Billa. Among those, ‘Food’ and ‘Home and
Hobby Utensils’ are the largest groups in both shops, followed by ‘Hygiene, Cosmetics
and Cloths’. The shares of the single product categories in relation to the total number
of products offered in the online shops closely mirror their shares in the German CPI
basket used to compute the official inflation rate (first column). All in all, about 25% of
German households’ total consumption expenditures are spent on the products offered in
the online shops.
The distribution of prices is shown in Figure (1). Separate density estimates are pro-
vided for products from the REWE and Billa online shop, covering only the period before
the VAT change and those products that are included in the harmonized product samples
from Table (1) above. While most items in both shops cost between one and five Euros,
the assortment of REWE also comprises higher priced products. This is also reflected in
the average product price: while the average product costs about four Euros at Billa, it is
17.5 Euros at REWE.
3. Results on the Pass-Through of the Temporary VAT Reduction
To get a first idea about the price dynamics around the announcement and implemen-
tation day of the VAT reduction, Figure (2) displays the share of REWE products whose
prices were lower (green shaded area), higher (red), or the same (grey) on a certain day
than they were on 2 May. While there do not appear to be visible price adjustments around
the announcement day on 3 June (dashed vertical line) or immediately afterwards, a week
of price increases followed only later starting in mid-June. The most notable changes in
prices can be observed, not surprisingly, around the date of the actual VAT change on 1
July (solid vertical line). Specifically, we can see that, beginning in the last week of June,
the share of products whose prices are lower than on 2 May increases abruptly from 15%
to around 30% within only a few days, with the largest number of price adjustments hap-
pening on 1 July. What is more, there does not appear to be a contemporaneous increase
in prices of other products.
What we cannot infer from Figure (2) is the magnitude of the price effect of the tem-
porary VAT reduction. To estimate this magnitude, we compare the development of the
supermarket prices in Germany to the development of the prices of the Billa supermarket
in Austria. The Austrian prices serve as a counterfactual, i.e. they show us how super-
market prices in Germany would have developed without the VAT reduction. Austrian
supermarket prices represent a particularly suitable counterfactual for our analysis for at
least three reasons. First, the dynamics of the Corona pandemic as well as the measures
implemented by the government to contain the spread of the Corona virus have been
comparable in Germany and Austria. Second, the Austrian government adopted a fiscal
stimulus package to combat the economic consequences of the Corona pandemic that is
5Fuest, Neumeier and Stöhlker
Figure 1: Comparison of Prices Levels
Probability density
.25
REWE
Billa
.2 Means
.15
.1
.05
0
0 10 20 30
Price (in €)
Notes: The density plot is based on price data collected between 4 June and 30 June, i.e. after the announce-
ment but before the VAT was changed.
very similar with regard to the measures and (relative) magnitude to the stimulus package
implemented in Germany. However, Austria did not adopt a general VAT rate cut.3 Third,
the Austrian stimulus package was announced less than three weeks after the German
one.
Figure (3) compares the development of daily retail prices across Germany and Aus-
tria. To facilitate the comparison, we transform the price for each single good into an
index value, the basis of the index being the product’s price on 4 June. Figure (3) reveals
that supermarket prices in Germany rose shortly before the VAT reduction. The increase
amounted to roughly 0.5%. However, as this price increase was also observed in Austria, it
is obviously independent of the VAT cut. In other words, the price increase in supermar-
ket retail in mid-June is not due to the anticipation of the VAT reduction. The comparison
with the data from Austria shows how important it is to identify counterfactual price de-
velopments. Without the comparison with Austria, it might have been concluded that
3 Austria only reduced the VAT rate for hotel stays as well as for food and drinks served in hotels.
6Price Pass-Through of Temporary VAT Cuts
Figure 2: Dynamics of Retail Sales Prices since May 2, 2020
Share of 116k REWE Products from May 2, 2020 (in %)
100
75
50
25
0
May 1 July 1 September 1 November 1
Price Reduced Price Increased Price Unchanged
Notes: The dashed, vertical line corresponds to June 3rd, i.e. the date of the announcement, and the solid,
vertical line denotes July 1st, the day of the VAT cut.
before the VAT reduction, prices were increased in Germany with the intention of only
giving the impression that the VAT reduction was passed on to the consumers.
Shortly after the price increase observed in both Germany and Austria, the (relative)
prices in German and Austrian supermarkets are drifting apart. Prices start to diverge
already at the end of June. This means that supermarket prices in Germany were already
being cut before the VAT reduction actually became effective. Until November, the (rela-
tive) supermarket prices in Germany were notably lower than the prices in Austria.
Note that in the weeks before and after the VAT reduction, the price trends in Austria
almost perfectly match the German trend, suggesting that Austrian supermarket prices
indeed represent an ideal counterfactual for the German prices.
To estimate the magnitude of the VAT reduction’s price effect and to test its signif-
icance, we use an event study approach. This approach also allows us to see how the
price effect evolves over time and provides an easy and intuitive way to verify whether
the common trends assumption holds (Schmidheiny and Siegloch, 2020). Specifically, we
7Fuest, Neumeier and Stöhlker
estimate the following empirical model:
17
X
piw = βj · bjiw + µi + θw + εiw ,
j=−3
Index i refers to the product, index w to the week of the observation and superscript j de-
notes the number of weeks until/after the VAT reduction. piw is the average price index of
product i in week w, bjiw is an indicator variable which is equal to one j weeks before/after
the VAT reduction in case product i belongs to the assortment of REWE and not to Billa
products, µi is a product-fixed effect that accounts for various time-invariant character-
istics, and εiw is the residual error term. Intuitively, this approach provides estimates for
week-specific mark-ups of REWE product prices over Billa prices. The corresponding es-
timates, together with their standard errors, are the coefficients βj with j denoting the
respective week before/after the implementation of the VAT reduction. Our event win-
dow covers 20 weeks, that is, the three weeks before the VAT reduction and the 17 weeks
afterwards. Since Figure (3) suggests that German supermarket prices already began to
decrease the week before the implementation, we use the fourth week of June 2020 as our
reference period.
Figure (4) plots the coefficient estimates for the event study indicators for each week
together along with the 99% confidence intervals. The results suggest that on average,
(relative) prices in Germany have fallen by about 2% as a result of the VAT reduction. This
indicates that the reduction in VAT has been almost completely passed on to consumers.4
4. The Role of Competition for VAT Pass-Through
From a theoretical perspective, the relationship between the intensity of competition in a
product market and the pass-through of commodity taxes to consumer prices is in general
ambiguous. In order to see this, consider the incidence of a tax τ per quantity of a traded
good in the two polar cases of perfect competition and monopoly. Suppose that the tax is
paid by the suppliers. Starting from the market clearing condition S(p − τ ) = D(p), the
pass-through of the tax under perfect competition is given by
dp 1
= εD p ,
dτ 1 − εS (p−τ )
p dS (p−τ )
where p is the consumer price and εD = − dD dp D
and εS = − d(p−τ ) S
are the elasticities
of demand and supply, respectively. The share of the tax burden that falls on consumers is
higher, the higher the elasticity of supply relative to the elasticity of demand. For example,
4 For products subject to the standard VAT rate, a full pass-through implies a price decrease of approxi-
mately 2.5%. For products taxed at the reduced rate, the price reduction in case of a full pass-through
would be approximately 1.9%. Our data set includes both products taxed at the regular rate and those
taxed at the reduced rate.
8Price Pass-Through of Temporary VAT Cuts
Figure 3: Comparison of Price Indexes at REWE and Billa
1-week mov. av. of Price Indices (June 4, 2020 = 100)
102
REWE (60k products)
Billa (8k products)
101
100
99
98
May 1 July 1 September 1 November 1
Notes: The dashed, vertical line corresponds to June 3rd, i.e. the date of the announcement, and the solid,
vertical line denotes July 1st, the day of the VAT cut.
if demand is perfectly inelastic, the incidence falls fully on the consumer. On the contrary,
if the demand elasticity becomes infinitely large, the incidence of the tax is borne by the
suppliers. Similarly, in case of perfectly elastic supply the pass-through of the tax to
consumer prices converges to unity and the tax incidence falls on the consumers.
What determines tax the incidence in imperfectly competitive markets? In the extreme
case of a perfectly price discriminating monopolist, the full burden of the tax is borne by
the monopolist because by definition prices are always set so that all buyers are reduced
to their reservation utility. In more realistic cases of imperfect competition, the derivation
of the incidence formula is slightly more complicated. Denote the quantity traded in the
market by q, the inverse demand function by p(q) and the marginal costs excluding taxes
by mc(q). Note that the the loss in consumer surplus (utility measured in terms of the
numeraire good) caused by a marginal decline q can be expressed as −p0 (q)q. Denote this
marginal loss by ms(q). Without price discrimination a profit maximizing monopolist
equates marginal revenue and marginal costs:
p(q) + p0 (q)q = mc(q) + τ. (1)
9Fuest, Neumeier and Stöhlker
Figure 4: Estimation Results of Event Study Design: The Impact of VAT Cut on Prices
Event Study Design: Effect on Price Index
1
99% Confidence Interval
Treatment Estimate
0
-1
-2
-3
-5 0 5 10 15 20
Weeks since 4th Week of June 2020
Using ms(q) = −p0 (q)q and εS = q·mc
mc
0 , differentiating (1) with respect to τ and rearrang-
ing yields
dp 1
= εD −1 1
,
dτ 1 + εS + εms
ms
where εms = q·ms 0 is the elasticity of the marginal surplus function. This elasticity mea-
sures the curvature of the logarithm of demand and can be positive or negative (Weyl and
Fabinger, 2013). Thus, from a theoretical perspective, it is ambiguous whether the pass-
through is higher or lower under monopoly than under perfect competition. For the case
of imperfect competition with more than one supplier, things are even more complicated.
Weyl and Fabinger (2013) provide an expression for the pass-through for the case with n
symmetrically differentiated firms. Let θ = p−mc(q)
p
εD be a parameter that describes the
degree of market power (reflected in price markups) with θ = 0 in the case of perfect
competition and θ = 1 in a monopolistic market. Weyl and Fabinger (2013) show that the
pass-through can be described as
dp 1
= θ εD −θ θ
, (2)
dτ 1+ εθ
+ εS
+ εms
10Price Pass-Through of Temporary VAT Cuts
where the last two terms of the denominator reflect the relative weights of perfect com-
petition and monopoly reaction according to the market’s competitive intensity. The sec-
ond term of the denominator sets θ in relation to its elasticity of quantity produced, i.e.
εθ = q θdθ .
dq
Without imposing further assumptions on the sign and the magnitude of its compo-
nents, the effect of competition on the extent of pass-through remains ambiguous. It is
helpful to discuss specific cases, which simplify the above equation, in order to better un-
derstand the role of the different elements and the impact of competition on pass-through.
Let us start with the term εDεS−θ which links elasticities of demand and supply with the
pass-through. Suppose that the marginal cost of supplying an additional product in the
online shop are constant, an assumption that is not unreasonable to impose at least in the
short-run. In this case the supply curve is horizontal and εS → ∞ so that the whole term
converges to zero. Likewise, if the degree of market power θ is constant in q then dθ dq
=0
and the second term of the denominator is zero. Assuming the extent of competition to be
constant is a common assumption in most empirical settings. For example, it is equal to
1 in the case of a monopolist, 0 in the case of perfect competition and in Bertrand models
of competition, and 1/n in the case of Cournot models, with n being the number of firms
operating in the market. As shown in Genakos and Pagliero (2019), price competition with
symmetrically differentiated products and linear demand functions also yields a constant
value for θ.
What remains to be discussed is the curvature of the demand function, i.e. the last
element of the denominator. It is not uncommon to assume that the demand function is
linear, which implies that εms = 1, in which case the pass-through of tax change to retail
prices increases with growing competitive pressure (declining θ) : dp/dτ = 1/(1+θ). This
is compatible with our empirical findings, as will be shown further below. Of course, in
different contexts, depending on the market-specificalities, other assumptions concerning
the functional form of the demand curve may be more appropriate, eventually leading to
different predictions regarding the impact of competition on the price pass-through of
taxes.
Figure 5 illustrates the relationship between the degree of competition, measured by
the number of brands in specific product sub-groups of the REWE assortment, and the
average price change within these product sub-groups since end-June for various points
in time over the post-treatment period. Every dot in each of these sub-figures corresponds
to a specific product sub-group, e.g. ‘Peanuts’, ‘Cocoa Powder’ or ‘Cigarettes’ with the
number of unique brands on the horizontal axis and the price index (prices on 4 June
equal 100) on the vertical axis. I.e., if the price index is below 100 for a specific sub-group
of products, then prices have declined, at least on average, while index values above 100
indicate price hikes since 4 June.
Starting with the sub-figure in the top left corner, the graph plots the distribution of
price indexes for all product groups two days after the reform implementation, i.e. on 2
July. Fitting a linear curve through all points indicates that prices have declined, as can
be seen from the fact that the linear curve is below the zero-line, but there is no clear
11Price Pass-Through of Temporary VAT Cuts
association between the magnitude of the price decrease and the degree of competition
within each product sub-group. Both the coefficient estimate as well as the corresponding
p-value from the linear fit do not indicate economic or statistical significance. As time
goes by, however, a relatively clear and robust pattern emerges from the figures: While
prices in product sub-groups with only a few brands have hardly changed since 4 June
(the intercept of the linear fit with the vertical axis is always close to 100), prices have
decreased visibly stronger among product groups with more product suppliers, indicating
a positive relationship between the degree of competition in a market and the price effect
of the VAT reduction.
5. Conclusion
On 3 June, the German federal government announced an unprecedented large fiscal stim-
ulus package to combat the economic consequences of the Corona pandemic. The most
important measure of that package was a temporary reduction of VAT rates. We study the
pass-through of this VAT tax cut on German supermarket retail prices using an extensive
webscraped data set covering on average 190,000 product prices per day.
The results of our analysis suggest that the temporary VAT reduction in Germany
was almost entirely passed on to consumers in supermarket retail. The product groups
covered by our analysis represent roughly 25% of overall private household consumption
spending. On average, we observe a decline in prices of around 2% in response to the
VAT rate cut. This is surprising inasmuch as studies on other temporary VAT reductions
conclude that the tax cut was only partially passed on and that the price effect became
smaller over time (Crossley et al., 2014). Studies on VAT reductions limited to specific
sectors suggest that the impact on consumer prices was very small, i.e., that the tax cuts
largely benefited firms (Benzarti and Carloni, 2019; Harju et al., 2018).
It remains to be seen how prices will develop once VAT rates return to their original
level. Experience from other countries shows that if VAT rates are only temporarily re-
duced, there is even a possibility that consumers will have to bear a greater burden in the
medium term because the price increase due to the return to the original VAT rate will be
higher than the initial price reduction resulting from the temporary VAT reduction (see
Benzarti et al., 2020; Benzarti and Carloni, 2019; Crossley et al., 2014). We can therefore
be curious to see how supermarket prices will develop in Germany from January 2021
onward.Fuest, Neumeier and Stöhlker
Figure 5: The Pass Through of the VAT Cut as a Function of Competitive Pressure
Average Price Index 0 Week(s) after Reform Implementation by Sub-Sub-Category Average Price Index 1 Week(s) after Reform Implementation by Sub-Sub-Category
104 104
Slope: -0.064 Slope: -0.148
(p-value: 0.3775) (p-value: 0.0665)
100 100
96 96
0 1 2 3 4 5 0 1 2 3 4 5
(Logarithm) Number of Brands in Sub-Sub-Category (Logarithm) Number of Brands in Sub-Sub-Category
Average Price Index 2 Week(s) after Reform Implementation by Sub-Sub-Category Average Price Index 4 Week(s) after Reform Implementation by Sub-Sub-Category
104 104
Slope: -0.199 Slope: -0.358
(p-value: 0.0190) (p-value: 0.0000)
100 100
96 96
0 1 2 3 4 5 0 1 2 3 4 5
(Logarithm) Number of Brands in Sub-Sub-Category (Logarithm) Number of Brands in Sub-Sub-Category
Average Price Index 8 Week(s) after Reform Implementation by Sub-Sub-Category Average Price Index 12 Week(s) after Reform Implementation by Sub-Sub-Category
104 104
Slope: -0.376 Slope: -0.416
(p-value: 0.0000) (p-value: 0.0000)
100 100
96 96
0 1 2 3 4 5 0 1 2 3 4 5
(Logarithm) Number of Brands in Sub-Sub-Category (Logarithm) Number of Brands in Sub-Sub-CategoryPrice Pass-Through of Temporary VAT Cuts References Benedek, D., De Mooij, R. A., Keen, M., and Wingender, P. Varieties of VAT Pass Through. International Tax and Public Finance, 27:890–930, 2019. Benzarti, Y. and Carloni, D. Who Really Benefits from Consumption Tax Cuts? Evidence from a Large VAT Reform in France. American Economic Journal: Economic Policy, 11 (1):38–63, 2019. Benzarti, Y., Carloni, D., Harju, J., and Kosonen, T. What Goes up May not Come Down: Asymmetric Incidence of Value-Added Taxes. Journal of Political Economy, 128(12), 2020. Blundell, R. Assessing the Temporary VAT Cut Policy in the UK. Fiscal Studies, 30(1): 31–38, 2009. Carbonnier, C. Who Pays Sales Taxes? Evidence from French VAT Reforms, 1987–1999. Journal of Public Economics, 91(5-6):1219–1229, 2007. Cole, A. J. Christmas in August: Prices and Quantities During a Sales Tax Holiday. PhD Dissertation, University of Michigan, 2009. Crossley, T. F., Low, H., and Wakefield, M. The Economics of a Temporary VAT Cut. Fiscal Studies, 30(1):3–16, 2009. Crossley, T. F., Low, H. W., and Sleeman, C. Using a Temporary Indirect Tax Cut as a Fiscal Stimulus: Evidence from the UK. IFS Working Papers, 2014. D‘Acunto, F., Hoang, D., and Weber, M. The Effect of Unconventional Fiscal Policy on Consumption Expenditure. NBER Working Paper 22563, 2016. D‘Acunto, F., Hoang, D., and Weber, M. Unconventional Fiscal Policy. In AEA Papers and Proceedings, volume 108, pages 519–23, 2018. Feldstein, M. The Role for Discretionary Fiscal Policy in a Low Interest Rate Environment. NBER Working Paper 9203, 2002. Genakos, C. and Pagliero, M. Competition and Pass-Through: Evidence from Isolated Markets. CEP WP 1638, 2019. Harju, J., Kosonen, T., and Skans, O. N. Firm Types, Price-setting Strategies, and Consumption-tax Incidence. Journal of Public Economics, 165:48–72, 2018. Harper, R. K., Hawkins, R. R., Martin, G. S., and Sjolander, R. Price Effects Around a Sales Tax Holiday: An Exploratory Study. Public Budgeting & Finance, 23(4):108–113, 2003. Hindriks, J. and Serse, V. The Incidence of VAT Reforms in Electricity Markets: Evidence from Belgium. CORE Discussion Paper 2020/12, 2020.
Fuest, Neumeier and Stöhlker Kosonen, T. More and Cheaper Haircuts after VAT Cut? On the Efficiency and Incidence of Service Sector Consumption Taxes. Journal of Public Economics, 131:87–100, 2015. Montag, F., Sagimuldina, A., and Schnitzer, M. Are Temporary Value-Added Tax Reduc- tions Passed on to Consumers? Evidence from Germany‘s Stimulus. arXiv preprint arXiv:2008.08511, 2020. Rickert, D., Schain, J. P., and Stiebale, J. Local Market Structure and Consumer Prices: Evidence from a Retail Merger. DICE Discussion Paper 280, 2018. Schmidheiny, K. and Siegloch, S. On Event Studies and Distributed-Lags in Two-way Fixed Effects Models: Identification, Equivalence, and Generalization. ZEW Discussion Paper 20-017, 2020. Weyl, E. G. and Fabinger, M. Pass-Through as an Economic Tool: Principles of Incidence under Imperfect Competition. Journal of Political Economy, 121(3):528–583, 2013.
You can also read
