33
Inflation with Covid Consumption Baskets Alberto Cavallo Harvard Business School & NBER May, 2020 Abstract The Covid-19 Pandemic has led to changes in consumer expenditure patterns that can introduce significant bias in the measurement of inflation. I use data collected from credit and debit transactions in the US to update the official basket weights and estimate the impact on the Consumer Price Index (CPI). I find that the Covid inflation rate is higher than the official CPI in the US, for both headline and core indices. I also find similar results with Covid baskets in 10 out of 16 additional countries. The difference is significant and growing over time, as social-distancing rules and behaviors are making consumers spend relatively more on food and other categories with rising inflation, and relatively less on transportation and other categories experiencing significant deflation. JEL-Codes: C43,E21,E31. Keywords: COVID, consumer expenditures, CPI, inflation. I am grateful to Florencia Hnilo for excellent research assistance, to John Friedman for sharing the Opportunity Insights data, to Caroline Coughlin and Manuel Bertolotto for assistance with the CPI data, and to Rafael Di Tella for helpful comments and suggestions.

In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

  • Upload
    others

  • View
    3

  • Download
    0

Embed Size (px)

Citation preview

Page 1: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

Inflation with Covid Consumption Baskets*

Alberto CavalloHarvard Business School & NBER

May, 2020

Abstract

The Covid-19 Pandemic has led to changes in consumer expenditure patterns that

can introduce significant bias in the measurement of inflation. I use data collected from

credit and debit transactions in the US to update the official basket weights and estimate

the impact on the Consumer Price Index (CPI). I find that the Covid inflation rate is

higher than the official CPI in the US, for both headline and core indices. I also find

similar results with Covid baskets in 10 out of 16 additional countries. The difference is

significant and growing over time, as social-distancing rules and behaviors are making

consumers spend relatively more on food and other categories with rising inflation, and

relatively less on transportation and other categories experiencing significant deflation.

JEL-Codes: C43,E21,E31.

Keywords: COVID, consumer expenditures, CPI, inflation.

*I am grateful to Florencia Hnilo for excellent research assistance, to John Friedman for sharing the Opportunity

Insights data, to Caroline Coughlin and Manuel Bertolotto for assistance with the CPI data, and to Rafael Di Tella

for helpful comments and suggestions.

Page 2: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

1 Introduction

The Covid-19 Pandemic has led to lockdowns, mobility restrictions, and social-distancing rules

that are dramatically changing consumer expenditure patterns in many countries.1 In partic-

ular, consumers are spending less on transportation, hotels, restaurants, and recreation, while

expenditures on food and other groceries have increased in both absolute and relative terms.

These sudden changes in expenditure patterns can introduce significant biases in the Con-

sumer Price Indices (CPIs) used to measure inflation, as noted by Diewert and Fox (2020). Most

National Statistical Offices (NSOs) update their CPI expenditure weights once a year, often with

lagged expenditure data. The US Bureau of Labor Statistics (BLS), for example, updated the

weights in December 2019 using expenditure information collected back in 2017-2018.2 While

this practice may be reasonable in normal times, it makes inflation indices much harder to in-

terpret during the Pandemic, as recently noted by central bankers such as Tenreyro (2020) and

Lane (2020).

In this paper, I empirically study the impact that changes in expenditure patterns are having

on the measurement of CPI inflation in 17 countries. I use high-frequency estimates of spending

based on transactional data to build updated CPI weights and compute alternative “Covid

Basket” price indices.

I start with the US, where card-based expenditure data are publicly available as part of

the Opportunity Insights Economic Tracker at Harvard and Brown University, described in

Chetty et al. (2020). I first show that headline Covid inflation is higher than what the CPI

reflects. By April 2020, the annual inflation rate of the US Covid index was 1.06%, compared

to only 0.35% of the equivalent CPI (all-items, US city average, not seasonally adjusted). The

difference is significant and growing over time, as new social-distancing rules and preferences

prevent consumer spending in categories that are experiencing deflation, such as transportation,

and induce more expenditure in food and other groceries, where prices are increasing over time.

I also find similar results when I focus on the Core CPI, implying that the bias is not limited to

1See Chetty et al. (2020), Carvalho et al. (2020), Baker et al. (2020), Andersen et al. (2020), Dunn et al. (2020),and Coibion et al. (2020).

2Most NSOs compute a Lowe Index formula at upper levels of aggregation. This introduces small adjustmentsthat account for relative price changes across categories every month, but they have little impact on the basketweights because quantities are assumed to be fixed. See Bureau of Labor Statistics (2020a) for an archive of BLSweights over time.

1

Page 3: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

the behavior of food and energy.

Next, I estimate the impact in 16 additional countries. To build a unique Covid basket for

each country, I update the official CPI weights with the same relative changes across categories

observed in the US. Consistent with the previous results, I find that in 10 countries the Covid

inflation rate is higher than that of the official CPI. In the other 6 cases, however, the inflation

rate is lower, highlighting the fact that the CPI bias can go in any direction, depending not only

on the changes in the basket weights but also on the inflation rates experienced by each CPI

subcategory.

My results have important implications for policy-makers trying to respond to the crisis.

They suggest that the cost of living for the average consumer is higher than estimated by the

official CPI, implying that real consumption is falling more quickly over time. The welfare effects

are particularly relevant during this crisis, where a large share of the population is becoming

unemployed. My results also highlight the divergence in sectoral inflation rates that reflect

different demand and supply shocks across categories in many countries. This particular behavior

of inflation during the Pandemic tends to amplify the CPI basket bias identified in this paper.

2 Data and Methodology

To build the “Covid Basket,” I start with daily measures of the change in US consumption pat-

terns since January 2020, available at the Opportunity Insights (OI) Tracker3. These estimates,

shown in Figure 1, are produced using transactional data collected from credit and debit card

transactions in the US, as described by Chetty et al. (2020).

3See https://opportunityinsights.org/

2

Page 4: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

-100

-50

050

100

Perc

ent C

hang

e

Feb-2020 Mar-2020 Apr-2020 May-2020

Grocery Apparel and Gen. Merch.Health care TransportationEntertainment and Recreation Restaurants and HotelsTotal

Figure 1: Changes in US Consumer Spending During the Pandemic

Notes: This graph shows the cumulative expenditure change across categories of goods and services in the US

since January 2020. These estimates are publicly available on the Opportunity Insights (OI) website, https:

//opportunityinsights.org/, and are produced using data collected from credit and debit card transactions by

Chetty et al. (2020).

I combine these estimates with official CPI data from the official NSO in each country,

including the Bureau of Labor Statistics in the US. In most cases, I use the CPI sector series

that form the first level of disaggregation of the headline CPI (all-items, not-seasonally adjusted),

as well as the latest available weights in the official CPI basket.

The matching of the OI categories with the CPI sectors requires some assumptions. Ta-

ble 1 shows the correspondence for the US data. To improve the matching, I split the CPI

for “Food and Beverages” into three additional subcategories. About six categories are then

closely matched in both datasets. For “Food at Home” and “Alcoholic Beverages,” I use the

OI “Grocery” category. For “Food Away from Home,” I use the OI category for “Restaurants

and Hotels.” For “Other Goods and Services,” I assume that the expenditure changes are equal

to those of the whole OI basket (“Total”). Finally, for “Housing” and “Education and Com-

munication,” I assume that expenditures in these categories have not changed, which seems a

reasonable assumption during the first months of the Pandemic.

3

Page 5: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

OI Expenditure ChangeUS CPI Category OI Expenditure Category (April 2020 monthly average)

Food at Home Grocery 12.3Alcoholic Beverages* Grocery 12.3Apparel Apparel and General Merchandise -41.0Housing** - 0.0Medical Care Health Care -52.2Transportation Transportation -70.1Recreation Entertainment and Recreation -71.1Education and communication** - 0.0Food Away from Home Restaurants and Hotels -61.9Other Goods and Services*** Total -26.8

Table 1: Matching US CPI and Expenditure Categories

Notes: *Assumes the same expenditure change as “Groceries”. **Assumes no change in expenditures. ***Assumesthe same change as the total expenditure.

To estimate the expenditure shares in the Covid basket, I start with the latest official CPI

weights and multiply them by the average percentage change in the corresponding expenditure

category each month. The new weights are then re-computed as a share of the total, to account

for the fact that total expenditure is also falling over time.

Formally, the Covid weights are given by:

sit =P itQ

it∑

i PitQ

it

=si0∆ei∑i s

i0∆ei

(1)

where P it and Qi

t are the prices and quantities of CPI category i at time t, and ∆ei = P itQ

it

P i0Q

i0

is

the change in expenditure. Equation 1 highlights the fact that these are relative weights, so the

importance of a category in the basket can change even when its expenditure is not affected.

The aggregate CPI and Covid price index are computed using the weighted sum of the changes

in the CPI sectoral indices, using weights si0 and sit, respectively.

3 Impact on US Inflation

I start with the results for the US, where both CPI and Covid expenditure estimates are publicly

available. Figure 2 and Table 2 show the impact that the new weights have on both the monthly

and annual inflation rates of the US CPI.

4

Page 6: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

-0.5

00.

000.

50

Jan 2019 Jan 2020

CPI Covid CPI

(a) Monthly Inflation Rate (%)

0.50

1.00

1.50

2.00

2.50

Jan 2019 Jan 2020

CPI Covid CPI

(b) Annual Inflation Rate (%)

Figure 2: US CPI with Covid Expenditure Weights

Notes: These graphs show the all-items, US city average, not seasonally adjusted CPI, and an equivalent index

constructed using estimates of the consumption expenditure shares under lockdown. The vertical gray line marks

the start of the Covid Pandemic in January 2020.

The official all-items CPI was relatively stable in January and February but started showing

deflation in March and April. The Covid index had a similar trend, but the magnitude of the

decline was significantly smaller. In March, the Covid index fell only -0.12%, compared to a fall

of -0.22% in the CPI. In April, the difference became more evident, with the Covid CPI falling

by only -0.09%, compared to a fall of -0.69% in the CPI. The impact on the annual inflation rate

is also significant. By April, the Covid index had an annual inflation rate of 1.06% compared to

0.35% in the official CPI.

5

Page 7: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

Monthly Inflation Rate Annual Inflation Rate

CPI Covid CPI CPI Covid CPI

January 0.39 0.39 2.50 2.50February 0.27 0.28 2.35 2.35March -0.22 -0.12 1.56 1.67April -0.69 -0.09 0.35 1.06

Table 2: US Inflation Rates During the Covid Pandemic

Notes: This table shows the monthly and annual inflation rate in the all-items, US city average, not seasonallyadjusted CPI, and an equivalent index constructed using estimates of the consumption expenditure shares duringthe Covid Pandemic.

To illustrate what is driving these results, Table 3 shows the category weights and incidence

details for April. The second column has the monthly CPI sector inflation used in both the

official and Covid indices. The third and fourth columns show the CPI and Covid weights in

each category. Finally, the last two columns show the incidence that each category has on the

total monthly inflation rate. The incidence is simply the monthly inflation rate multiplied by the

weight, so that the sum of all the incidence numbers in the last column is equal to the -0.09%

monthly inflation rate for the Covid CPI.

Monthly Weight Incidence

CPI Category CPI Inflation CPI Covid CPI CPI Covid CPI

Food at Home 2.67 7.58 11.28 0.20 0.30Alcoholic Beverages 0.30 1.02 1.52 0.00 0.00Apparel -4.38 2.81 2.20 -0.12 -0.10Housing -0.03 42.11 55.80 -0.01 -0.02Medical Care 0.28 8.83 5.60 0.02 0.02Transportation -4.97 15.74 6.25 -0.78 -0.31Recreation -0.27 5.82 2.23 -0.02 -0.01Education and Communication 0.13 6.77 8.97 0.01 0.01Food Away from Home 0.15 6.19 3.13 0.01 0.00Other Goods and Services -0.04 3.13 3.03 0.00 0.00

Table 3: US CPI Weights and Incidence - April 2020

Notes: The CPI weight is the share of expenditure in a given category over total expenditures. Note that categories

that experience no change in spending over time can have higher Covid weights as a share of the decreasing total

expenditure basket. The incidence is the monthly inflation rate multiplied by the weight. The sum of all the category

incidence numbers is equal to the monthly inflation rate.

Table 3 shows that the US Covid inflation rate is higher because there is more weight in

categories that had a positive inflation rate, and less weight in categories experiencing significant

6

Page 8: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

deflation. In particular, the weight for “Food at Home” rose from 7.58% to 11.28%, increasing

the incidence of this category by 0.1%. At the same time, the weight for “Transportation” fell

from 15.74% to 6.25%, increasing the incidence on the total monthly inflation rate by about

0.47%. The weights of “Housing” and “Education and Communication” also rose significantly.

However, these two categories had little impact on Covid inflation so far because their sectoral

inflation rates were close to zero percent in April.

3.1 Core CPI

The Covid basket bias is also present in the core index that excludes food and energy, as shown

in Table 4. The Covid Core deflation in April was only half of that in the Core CPI, while the

annual inflation rate is at 1.73% compared to the 1.43% in the official Core index. To build

these indices, I dropped all food series and split the “Housing” and “Transportation” categories

to remove their energy components. I also made similar assumptions for the consumer spending

patterns at the category level, with details provided in the Appendix.

Monthly Inflation Rate Annual Inflation Rate

Core CPI Covid Core CPI Core CPI Covid Core CPI

January 0.40 0.40 2.27 2.27February 0.47 0.47 2.37 2.36March 0.02 0.04 2.10 2.12April -0.46 -0.21 1.43 1.71

Table 4: US Core Inflation Rates During the Covid Pandemic

Notes: This table shows the monthly and annual inflation rate in the all-items less food and energy, US city average,

not seasonally adjusted CPI and an equivalent index constructed using estimates of the consumption expenditure

shares during the Covid Pandemic.

The Covid Core CPI has more inflation because there is less weight in subcategories such as

“Public Transportation,” “New and Used Motor Vehicles,” and “Motor Vehicle Insurance,” all

of which are experiencing significant deflation.

3.2 Other Downward Biases

My findings explore what happens with the expenditure basket at upper levels of the CPI, but

there are related measurement challenges at lower levels of aggregation as well. Diewert and

Fox (2020) describe the disappearing goods bias, which occurs when some products’ prices are

7

Page 9: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

no longer available to construct elementary price indices at the most disaggregated level of the

CPI. For example, the share of products with missing prices in the US CPI rose from 14% in

April 2019 to 34% in April 2020. In part, this reflects the challenges of collecting data during

this period (the BLS suspended physical data collections in March). Still, some prices are likely

missing due to stock-outs that resulted from the surge in panic-buying and supply disruptions

caused by the Pandemic.4 As Diewert and Fox (2020) note, the out-of-stock products are likely

to have higher market-clearing prices than those for continuing goods, potentially introducing

an additional downward bias on the CPI that reinforces the results in my paper.

The Pandemic is also likely to introduce an outlet bias, as a large share of total spending

moves online. In Cavallo (2017), I showed that multi-channel retailers tend to have identical

prices offline and online, so the products bought in this type of retailer (often under the “pick

up in store” modality) are not likely affected. However, the use of online delivery platforms,

such as Instacart and Shipt in the US, has soared during the Pandemic. Most retailers in these

platforms tend to have higher prices than in their physical stores.5 If this is not accounted for in

the data collection methodology used by the NSO, the change in spending outlets could cause

another downward bias in the CPI.

4 Impact in other Countries

In this section, I show similar results in 16 other countries using the same methodology with two

differences. First, these countries use the COICOP consumption classification system, which is

different from the one applied by the BLS in the US.6 Still, the matching and assumptions are

similar to those used in Table 1, as shown in the Appendix. Second, I do not have updated

expenditure estimates in other countries, so I assume that the change in spending for each

category during Covid is the same as in the US. The Covid shares are different in every country

because I start with the official CPI weights and update them under the assumption that the

relative changes in spending are similar to those experienced in the US. This provides a rough

approximation to the potential impact that the Covid consumption basket has in each of these

countries.

4See Bureau of Labor Statistics (2020b)5See Instacart (2020) and Shipt (2020)6See UN (2018) “Classification of Individual Consumption According to Purpose (COICOP)” for details.

8

Page 10: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

Table 5 compares the CPI and Covid inflation. I focus on the monthly and annual inflation

for April, with detailed weights and estimates shown in the Appendix.

Monthly Inflation Annual Inflation

Country CPI Covid CPI CPI Covid CPI

Argentina 1.74 1.91 44.67 45.10Brazil -0.28 0.20 2.10 2.76Canada -0.63 -0.36 0.08 0.54Chile -0.14 0.06 3.32 3.66France -0.01 0.23 0.53 0.87Japan -0.07 0.08 0.08 0.21Korea -0.63 -0.44 -0.02 0.33Spain 0.39 0.48 0.00 0.01Turkey 0.78 1.08 11.02 11.59Uruguay 2.06 2.72 10.80 11.56US -0.69 -0.09 0.35 1.06

Germany 0.55 0.46 0.89 0.80Greece -0.14 -0.11 0.02 -0.26Ireland -0.42 -0.50 -0.35 -0.59Italy 0.42 0.22 1.25 0.76Netherlands 0.93 0.51 1.11 0.77UK -0.19 -0.28 0.87 0.81

Table 5: CPI and Covid Inflation Rates in April 2020

Notes: The top panel shows countries where the Covid inflation is higher than the fixed-basket CPI. The bottom panelshows countries where the Covid inflation is lower than the fixed-basket CPI. Covid inflation rates are constructedusing official CPI weights in each country updated by the relative changes across categories observed in US data.Details on the incidence of CPI categories on the monthly inflation rate in each country are shown in the Appendix.

In the top panel, I show those countries where the Covid Inflation rate is higher than that

of the fixed-basket official CPI. Consistent with the US results, in most of these countries the

higher Covid inflation rate is driven by an increase in spending in “Food and Beverages” and a

decrease in the weight of “Transportation.”

The bottom panel shows that some countries are experiencing less inflation with the Covid

basket. This highlights the fact that the bias can go in any direction because it depends not

only of the changes in the basket weights but also on the inflation rates in each CPI sector. In

Germany, for example, there is less Covid inflation because there is less weight on “Recreation and

Culture,” a category that had a surprising inflation rate of 4.23% during April. In Ireland, there

is less Covid inflation because there is more weight on “Housing,” which is having significant

9

Page 11: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

deflation. Something similar occurs in Greece and Italy, where there is also less weight on

“Restaurants and Hotels,” a category that is experiencing a surprising positive inflation rate in

both countries. In the Netherlands, there is also less weight on “Restaurants and Hotels” with

a positive inflation rate. Finally, the UK is the only country that is still experiencing deflation

in “Food and Beverages,” where the weight has increased significantly.

Overall, my results outside the US should be seen as rough approximations to the Covid

inflation rates. An improved set of estimates requires updated Covid expenditure data in each

country. Some NSOs have started to produce experimental indices with ad-hoc adjustments to

the baskets, as in ONS (2020). Future research could help expand these efforts and account for

detailed Covid expenditure estimates using transactional data sources in each country.

5 Conclusion

There is a growing awareness among academics, central bankers, and the financial media about

the challenges of measuring and interpreting inflation during the Pandemic.7 A major con-

cern is that consumption patterns are greatly impacted by the lockdowns and social-distancing

behaviors, introducing significant bias into the measurement of CPI inflation.

I use estimates of the changes in consumption expenditures, obtained from US credit and

debit card transactions by Chetty et al. (2020), to update the basket of CPI weights and study

the effect on US inflation. I also provide estimates for the potential impact in 16 other countries.

In most cases, I show that the Covid price index has more inflation than the official CPI. By

April 2020, for example, the annual US all-items inflation rate was 1.06% with the Covid basket

and only 0.35% with the official CPI weights. The difference is large and is growing over time,

as consumers spend more on food and similar categories experiencing inflation, and less on

transportation and related categories with significant deflation.

These results have important implications for policy-makers trying to respond to the crisis,

as they suggest that the cost of living for the average consumer is higher than implied by the

official CPI. The welfare implications are particularly relevant for people losing their jobs during

the Pandemic. To further understand this impact, future research could try to estimate the

Covid basket of different population groups, including low-income households.

7See Diewert and Fox (2020), Tenreyro (2020), Lane (2020), and Wolf (2020).

10

Page 12: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

References

Andersen, Asger Lau, Emil Toft Hansen, Niels Johannesen, and Adam Sheridan,

“Consumer Responses to the COVID-19 Crisis: Evidence from Bank Account Transaction

Data,” Available at SSRN 3609814, 2020.

Baker, Scott R, R.A. Farrokhnia, Steffen Meyer, Michaela Pagel, and Constantine

Yannelis, “How Does Household Spending Respond to an Epidemic? Consumption During

the 2020 COVID-19 Pandemic,” Working Paper 26949, National Bureau of Economic Research

April 2020.

Bureau of Labor Statistics, “Consumer Price Index Archived News Releases : U.S. Bureau

of Labor Statistics,” https://www.bls.gov/bls/news-release/cpi.htm May 2020.

, “Impact of the Coronavirus (COVID-19) Pandemic on Consumer Price Index Data for

April 2020 : U.S. Bureau of Labor Statistics,” https://www.bls.gov/bls/consumer-price-index-

covid19-impacts-april-2020.htm April 2020.

Carvalho, Vasco, Juan Ramon Garcıa, Stephen Hansen, Alvaro Ortiz, Tomasa Ro-

drigo, Jose V. Rodrıguez Mora, and Pep Ruiz, “Tracking the COVID-19 Crisis through

the Lens of 1.4 Billion Transactions,” April 2020.

Cavallo, Alberto, “Are Online and Offine Prices Similar? Evidence from Large Multi-Channel

Retailers,” American Economic Review, 2017, 107 (1).

Chetty, Raj, John N. Friedman, Nathaniel Hendren, and Michael Stepner, “Real-

Time Economics: A New Platform to Track the Impacts of COVID-19 on People, Businesses,

and Communities Using Private Sector Data,” 2020.

Coibion, Olivier, Yuriy Gorodnichenko, and Michael Weber, “The Cost of the Covid-19

Crisis: Lockdowns, Macroeconomic Expectations, and Consumer Spending,” Working Paper

27141, National Bureau of Economic Research May 2020.

Diewert, W. Erwin and Kevin J Fox, “Measuring Real Consumption and CPI Bias under

Lockdown Conditions,” NBER Working Paper 27144 May 2020.

11

Page 13: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

Dunn, Abe, Kyle Hood, and Alexander Driessen, “Measuring the Effects of the COVID-

19 Pandemic on Consumer Spending Using Card Transaction Data,” US Bureau of Economic

Analysis Working Paper WP2020-5, 2020.

Instacart, “Instacart Help Center - What Can We Do for You?,”

https://www.instacart.com/help/section/200758534 2020.

Lane, Timothy, “Policies for the Great Global Shutdown and Beyond,”

https://www.bankofcanada.ca/2020/05/policies-great-global-shutdown-and-beyond/ May

2020.

ONS, “Prices Economic Analysis, Quarterly - Office for National Statistics,”

https://www.ons.gov.uk/economy/inflationandpriceindices/articles/priceseconomicanalysisquarterly/may2020

May 2020.

Shipt, “Shipt — How Are Your Prices Determined?,” https://help.shipt.com/pricing/how-are-

your-prices-determined 2020.

Tenreyro, Silvana, “Monetary Policy during Pandemics: Inflatin before, during, and after

Covid-19,” in “Bank of England” April 2020.

UN, “UN Classification of Individual Consumption According to Purpose (COICOP),” 2018.

Wolf, Martin, “Inflation Figures Are about to Get Fuzzier,”

http://ftalphaville.ft.com/2020/04/27/1587998585000/Inflation-figures-are-about-to-get-

fuzzier/ May 2020.

12

Page 14: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

Online Appendix

“ Inflation with Covid Consumption Baskets”

By Alberto CavalloMay 2020

A Additional Core CPI Results

-0.5

00.

000.

50

January 2019 January 2020

Core CPI Covid Core CPI

(a) Monthly Inflation Rate (%)

1.50

2.00

2.50

January 2019 January 2020

Core CPI Covid Core CPI

(b) Annual Inflation Rate (%)

Figure A1: US Core CPI with Covid Expenditure Weights

Notes: These graphs show the all-items less food and energy, US city average, not seasonally adjusted CPI and an

equivalent index constructed using estimates of the consumption expenditure shares under lockdown. The vertical

gray line marks the start of the Covid Pandemic in January 2020.

B COICOP and OI Expenditure Categories

Table A2 shows the matching between the COICOP CPI categories used by all countries (except

the US), and the OI expenditure data categories.

This table is equivalent to Table 1 in the paper. All assumptions remain the same, with

the addition of category ”Furnishings, Household Equipment, and Routine Maintenance of the

House” where I assume that the expenditures changes are equal to those of the total OI basket,

given that it is composed of a variety of household items, textiles, and related goods.

13

Page 15: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

Table A1: US Core - Weight and Incidence in April 2020

CPI Series

ExpenditureChange(vs De-cember

2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

CovidCPI

weight(%)

CPI In-cidence

CovidCPI In-cidence

AdditionalInci-dence

Alcoholic beverages 12.32 .3 1.02 1.91 0 .01 0Apparel -41.04 -4.38 2.81 2.75 -.12 -.12 0Household furnishings and operations 0 .4 4.6 7.63 .02 .03 .01Shelter 0 -.03 33.16 54.99 -.01 -.02 -.01Water and sewer and trash collection services 0 .13 1.08 1.8 0 0 0Medical care -52.18 .28 8.83 7.01 .02 .02 -.01Motor vehicle fees -70.05 -1.06 .57 .28 -.01 0 0Motor vehicle insurance -70.05 -7.7 1.7 .84 -.13 -.07 .07Motor vehicle maintenance and repair -70.05 .08 1.08 .53 0 0 0Motor vehicle parts and equipment -70.05 -.53 .4 .2 0 0 0New and used motor vehicles -70.05 -.54 7.28 3.61 -.04 -.02 .02Public transportation -70.05 -7.39 1.27 .63 -.09 -.05 .05Recreation -71.14 -.27 5.82 2.79 -.02 -.01 .01Education and communication 0 .13 6.77 11.23 .01 .01 .01Other goods and services -26.79 -.04 3.13 3.8 0 0 0

OI Expenditure ChangeCOICOP Code COICOP Category OI Expenditure Category (April 2020 monthly average)

100 Food and Non-Alcoholic Beverages Grocery 12.3200 Alcoholic Beverages and Tobacco Grocery 12.3300 Clothing and Footwear Apparel and General Merchandise -41.0400 Housing, Water, Electricity, Gas and other Fuels - 0.0500 Furnishings, Household Equip. and Maint. of the House Total -26.8600 Health Health Care -52.2700 Transport Transportation -70.1800 Communications - 0.0900 Recreation and Culture Entertainment and Recreation -71.11000 Education - 0.01100 Restaurants and Hotels Restaurants and Hotels -61.91200 Miscellaneous Goods and Services Total -26.8

Table A2: Matching CPI and Expenditure Categories in other Countries

C Weight and Incidence Tables in Other Countries

foreach country in usa argentina brazil canada chile colombia france germany greece ireland italy

japan korea netherlands russia southafrica spain turkey uk uruguay

14

Page 16: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

Table A3: Argentina - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 3.21 24.7 44.96 .79 1.44 .65200 12.32 1.01 2.57 4.68 .03 .05 .02300 -41.04 1.4 8.34 7.97 .12 .11 -.01400 0 .21 8.26 13.39 .02 .03 .01500 -26.79 .96 6.84 8.11 .07 .08 .01600 -52.18 .85 3.8 2.95 .03 .03 -.01700 -70.05 1.86 9.22 4.48 .17 .08 -.09900 -71.14 2.07 5.75 2.69 .12 .06 -.061000 0 -2.18 0 0 0 0 01100 -61.86 1.4 7.68 4.75 .11 .07 -.041200 -26.79 -.38 5.09 6.04 -.02 -.02 0

Online Appendix

“ Inflation with Covid Consumption Baskets”

By Alberto CavalloMay 2020

A Additional Core CPI Results

15

Page 17: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

Table A4: Brazil - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 2.24 14.82 23.85 .33 .53 .2200 12.32 .21 1.34 2.16 0 0 0300 -41.04 .1 6.11 5.16 .01 .01 0400 0 -.1 13.89 19.91 -.01 -.02 -.01500 -26.79 -1.37 8.17 8.57 -.11 -.12 -.01600 -52.18 -.22 5.42 3.71 -.01 -.01 0700 -70.05 -2.66 19.78 8.49 -.53 -.23 .3800 0 -.2 4.85 6.96 -.01 -.01 0900 -71.14 -.69 4.41 1.82 -.03 -.01 .021000 0 0 3.62 5.19 0 0 01100 -61.86 .76 8.5 4.64 .06 .04 -.031200 -26.79 .19 9.09 9.54 .02 .02 0

-0.5

00.

000.

50

January 2019 January 2020

Core CPI Covid Core CPI

(a) Monthly Inflation Rate (%)

1.50

2.00

2.50

January 2019 January 2020

Core CPI Covid Core CPI

(b) Annual Inflation Rate (%)

Figure A1: US Core CPI with Covid Expenditure Weights

Notes: These graphs show the all-items less food and energy, US city average, not seasonally adjusted CPI and an

equivalent index constructed using estimates of the consumption expenditure shares under lockdown. The vertical

gray line marks the start of the Covid Pandemic in January 2020.

16

Page 18: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

Table A5: Canada - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 .79 11.32 17.51 .09 .14 .05200 12.32 .35 2.54 3.93 .01 .01 0300 -41.04 -5.86 4.86 3.95 -.28 -.23 .05400 0 -.34 25.88 35.64 -.09 -.12 -.03500 -26.79 .4 6.19 6.24 .03 .03 0600 -52.18 .53 2.79 1.84 .01 .01 -.01700 -70.05 -1.58 17.04 7.03 -.27 -.11 .16800 0 -.75 3.46 4.76 -.03 -.04 -.01900 -71.14 -1.08 7.42 2.95 -.08 -.03 .051000 0 0 2.85 3.92 0 0 01100 -61.86 -.38 6.72 3.53 -.03 -.01 .011200 -26.79 .04 8.63 8.7 0 0 0

Table A6: Chile - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 1 19.06 28.13 .19 .28 .09200 12.32 1.44 3.31 4.89 .05 .07 .02300 -41.04 -3.24 4.48 3.47 -.15 -.11 .03400 0 .03 13.83 18.17 0 .01 0500 -26.79 -.03 7.02 6.75 0 0 0600 -52.18 .41 6.44 4.05 .03 .02 -.01700 -70.05 -.85 14.47 5.7 -.12 -.05 .07800 0 -.69 5 6.57 -.03 -.05 -.01900 -71.14 -.04 6.76 2.56 0 0 01000 0 -.42 8.09 10.63 -.03 -.04 -.011100 -61.86 -.19 4.37 2.19 -.01 0 01200 -26.79 -.84 7.16 6.89 -.06 -.06 0

Table A7: Colombia - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

17

Page 19: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

Table A8: France - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 2.13 16 25.42 .34 .54 .2200 12.32 -.33 4.24 6.74 -.01 -.02 -.01300 -41.04 -.05 4.93 4.11 0 0 0400 0 -.61 15.8 22.34 -.1 -.14 -.04500 -26.79 .35 5.85 6.06 .02 .02 0600 -52.18 .46 4.46 3.02 .02 .01 -.01700 -70.05 -1.45 15.92 6.74 -.23 -.1 .13800 0 -2.65 3.18 4.5 -.08 -.12 -.03900 -71.14 .74 8.92 3.64 .07 .03 -.041000 0 .02 .38 .54 0 0 01100 -61.86 -.88 8.31 4.48 -.07 -.04 .031200 -26.79 .36 12 12.42 .04 .04 0

Table A9: Germany - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 .99 12.21 18.66 .12 .19 .06200 12.32 1.25 4.33 6.62 .05 .08 .03300 -41.04 -.94 5.16 4.14 -.05 -.04 .01400 0 .09 21.12 28.74 .02 .03 .01500 -26.79 .58 5.67 5.65 .03 .03 0600 -52.18 .28 5.51 3.58 .02 .01 -.01700 -70.05 -1.42 15.48 6.31 -.22 -.09 .13800 0 -.1 6.06 8.24 -.01 -.01 0900 -71.14 4.23 13.24 5.2 .56 .22 -.341000 0 -.1 1.12 1.52 0 0 01100 -61.86 .18 5.62 2.92 .01 .01 01200 -26.79 .37 8.46 8.43 .03 .03 0

18

Page 20: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

Table A10: Greece - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 .77 19.59 31.22 .15 .24 .09200 12.32 -.39 5.11 8.15 -.02 -.03 -.01300 -41.04 .42 4.37 3.65 .02 .02 0400 0 -2.05 9.56 13.56 -.2 -.28 -.08500 -26.79 -.54 3.69 3.83 -.02 -.02 0600 -52.18 -.14 5.16 3.5 -.01 0 0700 -70.05 -1.53 16.38 6.96 -.25 -.11 .14800 0 -.54 5.25 7.45 -.03 -.04 -.01900 -71.14 .33 3.96 1.62 .01 .01 -.011000 0 0 2.45 3.48 0 0 01100 -61.86 1.11 17.82 9.64 .2 .11 -.091200 -26.79 .08 6.66 6.92 .01 .01 0

Table A11: Ireland - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 .32 11.87 19.98 .04 .06 .03200 12.32 1.18 5.82 9.8 .07 .12 .05300 -41.04 -1.98 5.21 4.6 -.1 -.09 .01400 0 -2.67 12.58 18.85 -.34 -.5 -.17500 -26.79 -.83 5.28 5.8 -.04 -.05 0600 -52.18 .19 6.09 4.37 .01 .01 0700 -70.05 -.78 14 6.28 -.11 -.05 .06800 0 -.11 3.37 5.04 0 -.01 0900 -71.14 .1 7.56 3.27 .01 0 01000 0 0 2.18 3.26 0 0 01100 -61.86 .45 18.64 10.65 .08 .05 -.041200 -26.79 -.52 7.39 8.1 -.04 -.04 0

19

Page 21: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

Table A12: Italy - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 1.23 17.52 28.22 .22 .35 .13200 12.32 .27 3.4 5.48 .01 .01 .01300 -41.04 4.5 8.47 7.16 .38 .32 -.06400 0 -3.87 11.41 16.36 -.44 -.63 -.19500 -26.79 .3 7.7 8.09 .02 .02 0600 -52.18 .1 4.31 2.95 0 0 0700 -70.05 -.58 14.79 6.35 -.09 -.04 .05800 0 0 2.78 3.98 0 0 0900 -71.14 0 6.24 2.58 0 0 01000 0 0 1.29 1.85 0 0 01100 -61.86 2.5 12.21 6.68 .31 .17 -.141200 -26.79 .09 9.81 10.3 .01 .01 0

Table A13: Japan - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 .86 19.84 27.54 .17 .24 .07200 12.32 -.19 1.64 2.28 0 0 0300 -41.04 2.14 4.13 3.01 .09 .06 -.02400 0 -.15 28.32 35 -.04 -.05 -.01500 -26.79 .69 3.49 3.16 .02 .02 0600 -52.18 -.19 4.3 2.54 -.01 0 0700 -70.05 -2.48 8.39 3.11 -.21 -.08 .13800 0 -.33 5.79 7.16 -.02 -.02 0900 -71.14 0 6.85 2.44 0 0 01000 0 -1.69 3.17 3.92 -.05 -.07 -.011100 -61.86 -.09 6.34 2.99 -.01 0 01200 -26.79 -.1 7.58 6.86 -.01 -.01 0

20

Page 22: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

Table A14: Korea - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 .03 14.21 22.21 0 .01 0200 12.32 -.02 1.14 1.79 0 0 0300 -41.04 -.77 6.75 5.54 -.05 -.04 .01400 0 -.15 10.42 14.51 -.02 -.02 -.01500 -26.79 -.08 3.79 3.86 0 0 0600 -52.18 .09 6.46 4.3 .01 0 0700 -70.05 -3.45 11.79 4.91 -.41 -.17 .24800 0 .66 6.2 8.63 .04 .06 .02900 -71.14 -.12 5.49 2.2 -.01 0 01000 0 -1.51 11.74 16.34 -.18 -.25 -.071100 -61.86 .07 12.68 6.73 .01 0 01200 -26.79 -.29 8.8 8.97 -.03 -.03 0

Table A15: Netherlands - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 .66 14.59 23.6 .1 .16 .06200 12.32 -.55 3.38 5.47 -.02 -.03 -.01300 -41.04 -.01 6.14 5.21 0 0 0400 0 -.35 15.78 22.73 -.06 -.08 -.02500 -26.79 .13 6.54 6.89 .01 .01 0600 -52.18 .36 2.93 2.02 .01 .01 0700 -70.05 1.33 14.1 6.08 .19 .08 -.11800 0 -.73 3.91 5.63 -.03 -.04 -.01900 -71.14 .67 12.35 5.13 .08 .03 -.051000 0 0 .98 1.41 0 0 01100 -61.86 7 8.73 4.8 .61 .34 -.281200 -26.79 .35 10.46 11.03 .04 .04 0

Table A16: Russia - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

21

Page 23: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

Table A17: Southafrica - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

Table A18: Spain - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 1.41 19.17 29.29 .27 .41 .14200 12.32 .15 2.92 4.46 0 .01 0300 -41.04 8.99 6.54 5.25 .59 .47 -.12400 0 -1.98 13.07 17.79 -.26 -.35 -.09500 -26.79 .33 5.71 5.69 .02 .02 0600 -52.18 .04 3.84 2.5 0 0 0700 -70.05 -3.12 14.71 5.99 -.46 -.19 .27800 0 -.17 6.63 9.02 -.01 -.02 0900 -71.14 1.02 7.13 2.8 .07 .03 -.041000 0 .01 1.6 2.18 0 0 01100 -61.86 1.16 14.02 7.27 .16 .08 -.081200 -26.79 .15 7.8 7.77 .01 .01 0

Table A19: Turkey - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 2.53 21.77 32.94 .55 .83 .28200 12.32 .02 5.87 8.88 0 0 0300 -41.04 4.08 7.32 5.81 .3 .24 -.06400 0 -.31 14.85 20 -.05 -.06 -.02500 -26.79 1.09 7.72 7.61 .08 .08 0600 -52.18 .53 2.63 1.69 .01 .01 0700 -70.05 -1.83 16.31 6.58 -.3 -.12 .18800 0 -1.23 7.5 10.11 -.09 -.12 -.03900 -71.14 -.67 3.62 1.41 -.02 -.01 .011200 -26.79 4.66 5.04 4.97 .24 .23 0

22

Page 24: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

Table A20: Uk - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 -.29 10.3 17.66 -.03 -.05 -.02200 12.32 .69 4.3 7.37 .03 .05 .02300 -41.04 -1.68 6.1 5.49 -.1 -.09 .01400 0 -.47 11.8 18.02 -.06 -.08 -.03500 -26.79 -1.61 6.1 6.82 -.1 -.11 -.01600 -52.18 -.18 2.6 1.9 0 0 0700 -70.05 0 16 7.32 0 0 0800 0 .44 4.9 7.48 .02 .03 .01900 -71.14 .37 14.8 6.52 .05 .02 -.031000 0 0 2.2 3.36 0 0 01100 -61.86 .62 12.6 7.34 .08 .05 -.031200 -26.79 -.85 9.6 10.73 -.08 -.09 -.01

Table A21: Uruguay - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 4.44 26.06 37.62 1.16 1.67 .51200 12.32 .31 4.29 6.2 .01 .02 .01300 -41.04 .63 5.37 4.07 .03 .03 -.01400 0 4.43 13.68 17.58 .61 .78 .17500 -26.79 1.37 5.73 5.39 .08 .07 0600 -52.18 .63 7.71 4.74 .05 .03 -.02700 -70.05 .33 10.13 3.9 .03 .01 -.02800 0 5.84 3.36 4.32 .2 .25 .06900 -71.14 -.8 6.49 2.41 -.05 -.02 .031000 0 -4.81 3.14 4.04 -.15 -.19 -.041100 -61.86 .55 7.7 3.78 .04 .02 -.021200 -26.79 .91 6.33 5.96 .06 .05 0

23

Page 25: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

Table A1: US Core - Weight and Incidence in April 2020

CPI Series

ExpenditureChange(vs De-cember

2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

CovidCPI

weight(%)

CPI In-cidence

CovidCPI In-cidence

AdditionalInci-dence

Alcoholic beverages 12.32 .3 1.02 1.91 0 .01 0Apparel -41.04 -4.38 2.81 2.75 -.12 -.12 0Household furnishings and operations 0 .4 4.6 7.63 .02 .03 .01Shelter 0 -.03 33.16 54.99 -.01 -.02 -.01Water and sewer and trash collection services 0 .13 1.08 1.8 0 0 0Medical care -52.18 .28 8.83 7.01 .02 .02 -.01Motor vehicle fees -70.05 -1.06 .57 .28 -.01 0 0Motor vehicle insurance -70.05 -7.7 1.7 .84 -.13 -.07 .07Motor vehicle maintenance and repair -70.05 .08 1.08 .53 0 0 0Motor vehicle parts and equipment -70.05 -.53 .4 .2 0 0 0New and used motor vehicles -70.05 -.54 7.28 3.61 -.04 -.02 .02Public transportation -70.05 -7.39 1.27 .63 -.09 -.05 .05Recreation -71.14 -.27 5.82 2.79 -.02 -.01 .01Education and communication 0 .13 6.77 11.23 .01 .01 .01Other goods and services -26.79 -.04 3.13 3.8 0 0 0

B COICOP and OI Expenditure Categories

Table A2 shows the matching between the COICOP CPI categories used by all countries (except

the US), and the OI expenditure data categories.

OI Expenditure ChangeCOICOP Code COICOP Category OI Expenditure Category (April 2020 monthly average)

100 Food and Non-Alcoholic Beverages Grocery 12.3200 Alcoholic Beverages and Tobacco Grocery 12.3300 Clothing and Footwear Apparel and General Merchandise -41.0400 Housing, Water, Electricity, Gas and other Fuels - 0.0500 Furnishings, Household Equip. and Maint. of the House Total -26.8600 Health Health Care -52.2700 Transport Transportation -70.1800 Communications - 0.0900 Recreation and Culture Entertainment and Recreation -71.11000 Education - 0.01100 Restaurants and Hotels Restaurants and Hotels -61.91200 Miscellaneous Goods and Services Total -26.8

Table A2: Matching CPI and Expenditure Categories in other Countries

This table is equivalent to Table 1 in the paper. All assumptions remain the same, with

the addition of category ”Furnishings, Household Equipment, and Routine Maintenance of the

House” where I assume that the expenditures changes are equal to those of the total OI basket,

given that it is composed of a variety of household items, textiles, and related goods.

24

Page 26: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

C Weight and Incidence Tables in Other Countries

foreach country in usa argentina brazil canada chile colombia france germany greece ireland italy

japan korea netherlands russia southafrica spain turkey uk uruguay

Table A3: Argentina - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 3.21 24.7 44.96 .79 1.44 .65200 12.32 1.01 2.57 4.68 .03 .05 .02300 -41.04 1.4 8.34 7.97 .12 .11 -.01400 0 .21 8.26 13.39 .02 .03 .01500 -26.79 .96 6.84 8.11 .07 .08 .01600 -52.18 .85 3.8 2.95 .03 .03 -.01700 -70.05 1.86 9.22 4.48 .17 .08 -.09900 -71.14 2.07 5.75 2.69 .12 .06 -.061000 0 -2.18 0 0 0 0 01100 -61.86 1.4 7.68 4.75 .11 .07 -.041200 -26.79 -.38 5.09 6.04 -.02 -.02 0

Table A4: Brazil - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 2.24 14.82 23.85 .33 .53 .2200 12.32 .21 1.34 2.16 0 0 0300 -41.04 .1 6.11 5.16 .01 .01 0400 0 -.1 13.89 19.91 -.01 -.02 -.01500 -26.79 -1.37 8.17 8.57 -.11 -.12 -.01600 -52.18 -.22 5.42 3.71 -.01 -.01 0700 -70.05 -2.66 19.78 8.49 -.53 -.23 .3800 0 -.2 4.85 6.96 -.01 -.01 0900 -71.14 -.69 4.41 1.82 -.03 -.01 .021000 0 0 3.62 5.19 0 0 01100 -61.86 .76 8.5 4.64 .06 .04 -.031200 -26.79 .19 9.09 9.54 .02 .02 0

25

Page 27: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

Table A5: Canada - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 .79 11.32 17.51 .09 .14 .05200 12.32 .35 2.54 3.93 .01 .01 0300 -41.04 -5.86 4.86 3.95 -.28 -.23 .05400 0 -.34 25.88 35.64 -.09 -.12 -.03500 -26.79 .4 6.19 6.24 .03 .03 0600 -52.18 .53 2.79 1.84 .01 .01 -.01700 -70.05 -1.58 17.04 7.03 -.27 -.11 .16800 0 -.75 3.46 4.76 -.03 -.04 -.01900 -71.14 -1.08 7.42 2.95 -.08 -.03 .051000 0 0 2.85 3.92 0 0 01100 -61.86 -.38 6.72 3.53 -.03 -.01 .011200 -26.79 .04 8.63 8.7 0 0 0

Table A6: Chile - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 1 19.06 28.13 .19 .28 .09200 12.32 1.44 3.31 4.89 .05 .07 .02300 -41.04 -3.24 4.48 3.47 -.15 -.11 .03400 0 .03 13.83 18.17 0 .01 0500 -26.79 -.03 7.02 6.75 0 0 0600 -52.18 .41 6.44 4.05 .03 .02 -.01700 -70.05 -.85 14.47 5.7 -.12 -.05 .07800 0 -.69 5 6.57 -.03 -.05 -.01900 -71.14 -.04 6.76 2.56 0 0 01000 0 -.42 8.09 10.63 -.03 -.04 -.011100 -61.86 -.19 4.37 2.19 -.01 0 01200 -26.79 -.84 7.16 6.89 -.06 -.06 0

Table A7: Colombia - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

26

Page 28: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

Table A8: France - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 2.13 16 25.42 .34 .54 .2200 12.32 -.33 4.24 6.74 -.01 -.02 -.01300 -41.04 -.05 4.93 4.11 0 0 0400 0 -.61 15.8 22.34 -.1 -.14 -.04500 -26.79 .35 5.85 6.06 .02 .02 0600 -52.18 .46 4.46 3.02 .02 .01 -.01700 -70.05 -1.45 15.92 6.74 -.23 -.1 .13800 0 -2.65 3.18 4.5 -.08 -.12 -.03900 -71.14 .74 8.92 3.64 .07 .03 -.041000 0 .02 .38 .54 0 0 01100 -61.86 -.88 8.31 4.48 -.07 -.04 .031200 -26.79 .36 12 12.42 .04 .04 0

Table A9: Germany - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 .99 12.21 18.66 .12 .19 .06200 12.32 1.25 4.33 6.62 .05 .08 .03300 -41.04 -.94 5.16 4.14 -.05 -.04 .01400 0 .09 21.12 28.74 .02 .03 .01500 -26.79 .58 5.67 5.65 .03 .03 0600 -52.18 .28 5.51 3.58 .02 .01 -.01700 -70.05 -1.42 15.48 6.31 -.22 -.09 .13800 0 -.1 6.06 8.24 -.01 -.01 0900 -71.14 4.23 13.24 5.2 .56 .22 -.341000 0 -.1 1.12 1.52 0 0 01100 -61.86 .18 5.62 2.92 .01 .01 01200 -26.79 .37 8.46 8.43 .03 .03 0

27

Page 29: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

Table A10: Greece - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 .77 19.59 31.22 .15 .24 .09200 12.32 -.39 5.11 8.15 -.02 -.03 -.01300 -41.04 .42 4.37 3.65 .02 .02 0400 0 -2.05 9.56 13.56 -.2 -.28 -.08500 -26.79 -.54 3.69 3.83 -.02 -.02 0600 -52.18 -.14 5.16 3.5 -.01 0 0700 -70.05 -1.53 16.38 6.96 -.25 -.11 .14800 0 -.54 5.25 7.45 -.03 -.04 -.01900 -71.14 .33 3.96 1.62 .01 .01 -.011000 0 0 2.45 3.48 0 0 01100 -61.86 1.11 17.82 9.64 .2 .11 -.091200 -26.79 .08 6.66 6.92 .01 .01 0

Table A11: Ireland - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 .32 11.87 19.98 .04 .06 .03200 12.32 1.18 5.82 9.8 .07 .12 .05300 -41.04 -1.98 5.21 4.6 -.1 -.09 .01400 0 -2.67 12.58 18.85 -.34 -.5 -.17500 -26.79 -.83 5.28 5.8 -.04 -.05 0600 -52.18 .19 6.09 4.37 .01 .01 0700 -70.05 -.78 14 6.28 -.11 -.05 .06800 0 -.11 3.37 5.04 0 -.01 0900 -71.14 .1 7.56 3.27 .01 0 01000 0 0 2.18 3.26 0 0 01100 -61.86 .45 18.64 10.65 .08 .05 -.041200 -26.79 -.52 7.39 8.1 -.04 -.04 0

28

Page 30: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

Table A12: Italy - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 1.23 17.52 28.22 .22 .35 .13200 12.32 .27 3.4 5.48 .01 .01 .01300 -41.04 4.5 8.47 7.16 .38 .32 -.06400 0 -3.87 11.41 16.36 -.44 -.63 -.19500 -26.79 .3 7.7 8.09 .02 .02 0600 -52.18 .1 4.31 2.95 0 0 0700 -70.05 -.58 14.79 6.35 -.09 -.04 .05800 0 0 2.78 3.98 0 0 0900 -71.14 0 6.24 2.58 0 0 01000 0 0 1.29 1.85 0 0 01100 -61.86 2.5 12.21 6.68 .31 .17 -.141200 -26.79 .09 9.81 10.3 .01 .01 0

Table A13: Japan - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 .86 19.84 27.54 .17 .24 .07200 12.32 -.19 1.64 2.28 0 0 0300 -41.04 2.14 4.13 3.01 .09 .06 -.02400 0 -.15 28.32 35 -.04 -.05 -.01500 -26.79 .69 3.49 3.16 .02 .02 0600 -52.18 -.19 4.3 2.54 -.01 0 0700 -70.05 -2.48 8.39 3.11 -.21 -.08 .13800 0 -.33 5.79 7.16 -.02 -.02 0900 -71.14 0 6.85 2.44 0 0 01000 0 -1.69 3.17 3.92 -.05 -.07 -.011100 -61.86 -.09 6.34 2.99 -.01 0 01200 -26.79 -.1 7.58 6.86 -.01 -.01 0

29

Page 31: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

Table A14: Korea - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 .03 14.21 22.21 0 .01 0200 12.32 -.02 1.14 1.79 0 0 0300 -41.04 -.77 6.75 5.54 -.05 -.04 .01400 0 -.15 10.42 14.51 -.02 -.02 -.01500 -26.79 -.08 3.79 3.86 0 0 0600 -52.18 .09 6.46 4.3 .01 0 0700 -70.05 -3.45 11.79 4.91 -.41 -.17 .24800 0 .66 6.2 8.63 .04 .06 .02900 -71.14 -.12 5.49 2.2 -.01 0 01000 0 -1.51 11.74 16.34 -.18 -.25 -.071100 -61.86 .07 12.68 6.73 .01 0 01200 -26.79 -.29 8.8 8.97 -.03 -.03 0

Table A15: Netherlands - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 .66 14.59 23.6 .1 .16 .06200 12.32 -.55 3.38 5.47 -.02 -.03 -.01300 -41.04 -.01 6.14 5.21 0 0 0400 0 -.35 15.78 22.73 -.06 -.08 -.02500 -26.79 .13 6.54 6.89 .01 .01 0600 -52.18 .36 2.93 2.02 .01 .01 0700 -70.05 1.33 14.1 6.08 .19 .08 -.11800 0 -.73 3.91 5.63 -.03 -.04 -.01900 -71.14 .67 12.35 5.13 .08 .03 -.051000 0 0 .98 1.41 0 0 01100 -61.86 7 8.73 4.8 .61 .34 -.281200 -26.79 .35 10.46 11.03 .04 .04 0

Table A16: Russia - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

30

Page 32: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

Table A17: Southafrica - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

Table A18: Spain - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 1.41 19.17 29.29 .27 .41 .14200 12.32 .15 2.92 4.46 0 .01 0300 -41.04 8.99 6.54 5.25 .59 .47 -.12400 0 -1.98 13.07 17.79 -.26 -.35 -.09500 -26.79 .33 5.71 5.69 .02 .02 0600 -52.18 .04 3.84 2.5 0 0 0700 -70.05 -3.12 14.71 5.99 -.46 -.19 .27800 0 -.17 6.63 9.02 -.01 -.02 0900 -71.14 1.02 7.13 2.8 .07 .03 -.041000 0 .01 1.6 2.18 0 0 01100 -61.86 1.16 14.02 7.27 .16 .08 -.081200 -26.79 .15 7.8 7.77 .01 .01 0

Table A19: Turkey - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 2.53 21.77 32.94 .55 .83 .28200 12.32 .02 5.87 8.88 0 0 0300 -41.04 4.08 7.32 5.81 .3 .24 -.06400 0 -.31 14.85 20 -.05 -.06 -.02500 -26.79 1.09 7.72 7.61 .08 .08 0600 -52.18 .53 2.63 1.69 .01 .01 0700 -70.05 -1.83 16.31 6.58 -.3 -.12 .18800 0 -1.23 7.5 10.11 -.09 -.12 -.03900 -71.14 -.67 3.62 1.41 -.02 -.01 .011200 -26.79 4.66 5.04 4.97 .24 .23 0

31

Page 33: In ation with Covid Consumption Baskets › ... › Paper_Covid_Price.pdf · the impact on the Consumer Price Index (CPI). I nd that the Covid in ation rate is higher than the o cial

Table A20: Uk - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 -.29 10.3 17.66 -.03 -.05 -.02200 12.32 .69 4.3 7.37 .03 .05 .02300 -41.04 -1.68 6.1 5.49 -.1 -.09 .01400 0 -.47 11.8 18.02 -.06 -.08 -.03500 -26.79 -1.61 6.1 6.82 -.1 -.11 -.01600 -52.18 -.18 2.6 1.9 0 0 0700 -70.05 0 16 7.32 0 0 0800 0 .44 4.9 7.48 .02 .03 .01900 -71.14 .37 14.8 6.52 .05 .02 -.031000 0 0 2.2 3.36 0 0 01100 -61.86 .62 12.6 7.34 .08 .05 -.031200 -26.79 -.85 9.6 10.73 -.08 -.09 -.01

Table A21: Uruguay - Weight and Incidence in April 2020

COICOP Category

ExpenditureChange

(vsDecember2019, %)

CPIMonthlyInflation

Rate(4/2020,

%)

CPIweight

(%)

Covid CPIweight

(%)

CPIIncidence

Covid CPIIncidence

AdditionalIncidence

100 12.32 4.44 26.06 37.62 1.16 1.67 .51200 12.32 .31 4.29 6.2 .01 .02 .01300 -41.04 .63 5.37 4.07 .03 .03 -.01400 0 4.43 13.68 17.58 .61 .78 .17500 -26.79 1.37 5.73 5.39 .08 .07 0600 -52.18 .63 7.71 4.74 .05 .03 -.02700 -70.05 .33 10.13 3.9 .03 .01 -.02800 0 5.84 3.36 4.32 .2 .25 .06900 -71.14 -.8 6.49 2.41 -.05 -.02 .031000 0 -4.81 3.14 4.04 -.15 -.19 -.041100 -61.86 .55 7.7 3.78 .04 .02 -.021200 -26.79 .91 6.33 5.96 .06 .05 0

32