Combining administrative and market data in the development of new commercial real estate - PowerPoint Presentation

Slide1

Combining administrative and market data in the development of new commercial real estate indicators

Sigrid

Krogstrup Jensen, sij@dst.dk

Cajsa Mølskov, cms@dst.dk

Property that generates a profit for its owner, and thus, must be valuated and taxed accordingly

Business

Mixed housing and businessWarehouse and productionPrivate insitutionsSpecialized property

Commercial real estate in Denmark

2

According to the Danish Dwellings and Buildings register (BBR) a property consists of buildings and units:

Commercial real estate in

Denmark II3

Property

Building 2

Building 1

Unit 4

Unit 1

Unit 2

Unit 3

Unit 5

Number of commercial properties in Denmark

4

Business

Mixed housing and

business

Production and warehouse

Private institutions

Specialised property

Other

Total

Capital region

844

2.744

1.005

72

82

325

5.072

Sealand region

1.225

514

827

129

268

449

3.412

Southern

Denmark

region

1.822

735

1.171

189

595

1.037

5.549

Central

Jutland

region

1.813

1.054

1.709

130

732

743

6.181

Northern Jutland region

1.044

646

735

119

625

710

3.879

All of Denmark

6.748

5.693

5.447

639

2.302

3.264

24.093

Administrative data sources and data

5

The Danish FSA

The Danish Buildings and Property Agency

Landsbyggefonden

Rent (net/gross)

Market rent

X

Actual rent

X

X

X

Operating costs (net/gross)

Market operating costs

Actual operating costs

X

X

X

Yield

X

Location

Property ID (BBR)

X

X

X

Address

X

X

Type of property

X

X

X

Type of transaction

Sale

Rent

X

Date

X

X

X

M2

X

X

Value

Price

Valuation

X

Vacancy

Market data sources and data6

The Danish Property Federation

Ejendomstorvet

Lokalebasen

Rent (net/gross)

Market rent

X

X

X

Actual rent

 X

X

Operating costs (net/gross)

Market operating costs

X

X

Actual operating costs

X

X

Yield

 X

X

Location

Property ID (BBR)

X

Address

X

X

X

Type of property

X

X

X

Type of transaction

Sale

X

X

Rent

X

X

Date

X

X

X

M2

 X

X

X

Value

Price

X

X

Valuation

X

Vacancy

X 

Data processing7

InitializationAll raw data is stored in separate

tables

Duplicate records are removedAutomatic validation and calculationsLogical checks are performedRent per m2 is calculated

All corrections and calculations

are

flagged

Initialization and automatic validation and calculations

8

Why? Validation Enrichment of the data with administrative ID’s

How?

Address, area and property typeAdministrative keysIndicators of quality Automatic match to the BBR9

Addresses are firstly

validated

using Danish Addresses Web API (DAWA)DAWA uses data from the Danish Address Register (DAR)DAWA has been designed to service IT systems that use addressesOur system uses the ”Address cleaning” process

where an unstructured address is

translated

to a

correct

address

Automatic

match - DAWA

10

DAWA

assesses

the quality of the returned address ID; A, B, CAutomatic match – DAWA address cleaning11

Unstructured

a

ddress

Query

Valid

a

ddress

ID

BBR

DAWA checks

address

in DAR

Matches are validated by comparing

the m

2 and the type of property between the BBR and the received dataTypes of matches:Correct matchPreliminary matchMatched but flaggedMissed matchMatch quality:Good match Medium match

Inferior matchAutomatic

match – Match

validation

12

In the manual treatment there are three possible outcomes:The cause of the failure to match correctly is corrected and the matching process is repeated

The

data cannot be corrected but the match, however, is assesed to be correct and the match is forced through (forced match)The data cannot be corrected, the match is assesed to be incorrect and the match is given up (abandoned match)If there are no match variables available for the observation the match is given up automatically and the observation will not be treated manually.

Manual match and validation

13

Data source

Time period

No. of unique observations

Match percentages

 Administrative data

Unit level

Building level

Property level

No match

The Danish FSA

2012-2018

1.804.875

69,1

1,2

29,4

0,3

The Danish Building and

Property Agency

2017-2018

339

10,3

19,5

68,1

2,1

Landsbyggefonden

2013-2018

575.160

99,6

0,0

0,0

0,4

Market data

The Danish Property Federation

2017-2018

56.345

52,3

19,6

28,1

0,0

Ejendomstorvet

2014-2018

45.860

15,0

16,4

36,6

32,4

Lokalebasen - supply

2014-2018

106.794

14,7

17,5

44,3

23,5

Lokalebasen - realised

2014-2018

4.483

16,8

12,5

25,5

45,2

Preliminary match

results

14

AdvantagesEasy access to dataRegular

and

consistent data collectionTotal coverageHolds administrative keysDisadvantagesDoes not always cover the target populationCan in some cases only be

used for approximation

Administrative data for CREI-

production

15

AdvantagesDirectly reflects the

market

In some cases data collection can be ordered to ensure representativityDisadvantagesData is privately owned

and very sensitiveVariables can

be

inconsistent

within

the data

Only

smaller parts of

the population

is coveredData

rarely holds administrative keysMarket data for CREI-production

16

Commercial real estate indicators of the dynamics of supply and demand:

Commercial

property pricesRental prices on commercial propertyVacancy ratesCommercial property for rent or saleBuilding permits for commercial propertyLending supply and criteriaDemand for new indicators

17

CPPICommercial property for saleRent per

m

2 on housingIndicators to be substantiated18

Rental prices on commercial property

Vacancy

ratesCommercial property for rent Indicators to be developed19