E-Banking Snapshot 31
Innovations in P2P lending may put computers over people
November 2009
Digital economy and structural change
Author
Thomas Meyer
+49 69 910-46830
thomas-d.meyer@db.com
Editor
Antje Stobbe
Technical Assistant
Sabine Kaiser
Deutsche Bank Research
Frankfurt am Main
Germany
Internet:www.dbresearch.com
E-mail marketing.dbr@db.com
Fax: +49 69 910-31877
Managing Director
Norbert Walter
Welcome to the machine
P2P lending is an alternative way of originating loans (mostly) without
banks. Peer-to-peer (P2P) lending sites broker small loans between savers and
lenders online. The loans are usually unsecured so that lenders bear the full credit
risk. Since 2005, many P2P lending sites have popped up around the globe (see
E-Banking Snapshot 22 for more background). There are substantial differences
between the sites: Kiva, for instance, channels loans towards entrepreneurs in
developing countries, whereas other sites target lenders and borrowers in mature
markets. chart 1
The largest sites in the US have originated loans worth more than
USD 200 m in total. This is, however, only a tiny fraction of the overall volume
of outstanding consumer loans in the US – which stands at USD 2.4 trillion.
chart 2
High interest rates and media coverage attract lenders. Many German
lenders, for instance, flock to P2P lending because the promised interest rates
exceed that of, say, bank deposits. The flipside is of course, that loans are not
necessarily cheaper than those offered by traditional financial firms. chart 3
Lenders may underestimate risks. At Smava, a German P2P lending site,
lenders do not bear the full credit risk because losses (on the principal, not the
interest) are spread pro rata to all loans of the same credit grade. This encourages
risk taking. In fact, more than one-third of all loans at Smava are awarded to
business owners or self-employed professionals. Their credit grades may
underestimate the true risk because the grades may reflect personal credit history,
not that of the business. chart 4
Selection and pressure by peers lose importance. In the early days of
P2P lending, borrowers could join a group of like-minded borrowers and get
endorsements from friends as a way to signal quality and their commitment to
repaying the loan. Group members used to have lower default ratios. But peer
pressure seems to recede in mature markets: few sites even feature groups and
lenders tend to focus more on quantitative indicators instead. This year, only 13
new groups were created at Smava and they featured only one loan in total.
chart 5
Innovations in P2P lending shift power towards computers. Not all
lenders fancy reading all the loan bids. Therefore, some sites have introduced
automatic bidding and secondary markets. They allow lenders to bid on loan
requests automatically that match some pre-defined metrics and to trade existing
loan notes on dedicated platforms. This injects new professionalism into P2P
lending but also shifts power from people to computers. chart 6
Borrowers need to explain their situation. Borrowers at Lending Club use
on average 370 characters to describe their loan request. They write more if their
credit grade is bad. Each notch downward (there are 35 from A1 to G5) is
statistically associated with 3.5 additional characters. chart 7
Ignore explanations at your own peril. Lengthy explanations tend to
increase the chances of getting funded but are also associated with a higher
delinquency risk – other things being equal. As it appears, borrowers with a more
complicated background need to explain that at length to prospective lenders who
can then make informed choices. Lenders who ignore this information might load
up too many bad loans during this downturn. chart 8
E-Banking Snapshot 31 Welcome to the machine

P2P lending is an alternative way of

Varieties of P2P lending originating loans (mostly) without
Examples banks. Peer-to-peer (P2P) lending sites
Zopa Prosper Kiva Sm ava broker small loans between savers and
UK lenders online. The loans are usually
Region US Global DE unsecured so that lenders bear the full
(US, IT, JP)*
Poverty credit risk. Since 2005, many P2P
Purpose Profit Profit Profit lending sites have popped up around
alleviation
Listings & the globe (see E-Banking Snapshot 22
Loans Listings Listings Listings
matching for more background). There are
Credit risk Lender Lender Lender Lender pool substantial differences between the
Auction & Set by sites: Kiva, for instance, channels loans
Interest rate Auction None towards entrepreneurs in developing
matching borrow er
Secondary countries, whereas other sites target
No Yes No No
trading lenders and borrowers in mature
Autom atic markets. back to front page
No Yes No Yes
bidding

*planned. Source: DB Research, 2009 1

The largest sites in the US have

Out of kindergarten originated loans worth more than USD
Total value of originated loans since inception at selected P2P lending sites 200 m in total. This is, however, only a
(USD m)
tiny fraction of the overall volume of
Prosper (US) outstanding consumer loans in the US
– which stands at USD 2.4 trillion. back
Kiva (micro-finance) to front page

Lending Club (US)

Smava (DE)

0 50 100 150 200

Zopa does not disclose lending volumes. Sources: Company websites, as of 29 Sep 2009 2

High interest rates and media coverage

Not necessarily cheaper attract lenders. Many German lenders,
Effective rate of interest on a EUR 10.000 unsecured loan, 36 months for instance, flock to P2P lending
duration, medium credit quality in Germany (selected offers)
because the promised interest rates
exceed that of, say, bank deposits. The
Postbank flipside is of course, that loans are not
Citibank necessarily cheaper than those offered
Smava* by traditional financial firms. back to
ING-DiBa front page
netbank
norisbank
CosmosDirekt
1822direkt

0 2 4 6 8 10 12

*expected As of 29 Sep 2009. Source: FMH Finanzberatung, 2009 3

November 2009 2
E-Banking Snapshot 31 Welcome to the machine

Lenders may underestimate risks. At

Many loans to business owners Smava, a German P2P lending site,
Loans by occupation of the borrower at Smava (DE), %
lenders do not bear the full credit risk
because losses (on the principal, not
Public servants the interest) are spread pro rata to all
Business
owners*
4% loans of the same credit grade. This
35% encourages risk taking. In fact, more
than one-third of all loans at Smava are
awarded to business owners or self-
Employees
51%
employed professionals. Their credit
Pensioners grades may underestimate the true risk
10% because the grades may reflect
personal credit history, not that of the
* Include self-employed, entrepreneurs and owner-managers business. back to front page
Sources: DB Research, Smava, Wiseclerk.com, 2009 4

Selection and pressure by peers lose

Fewer new groups importance. In the early days of P2P
Number of groups at Smava (DE), by founding year lending, borrowers could join a group of
60 like-minded borrowers and get
48 endorsements from friends as a way to
50
38 signal quality and their commitment to
40 repaying the loan. Group members
30 used to have lower default ratios. But
peer pressure seems to recede in
13 20
mature markets: few sites even feature
10 groups and lenders tend to focus more
0
on quantitative indicators instead. This
year, only 13 new groups were created
2007 2008 2009*
at Smava and they featured only one
* Until October 2009 Sources: DB Research, Smava, 2009 5 loan in total. back to front page

Innovations in P2P lending shift power

Professional tools rather than peer pressure towards computers. Not all lenders
Major developments in P2P lending fancy reading all the loan bids.
Automated bidding allows to generate bids according to Therefore, some sites have introduced
pre-defined metrics: Personal impressions derived from loan automatic bidding and secondary
descriptions do not count. markets. They allow lenders to bid on
loan requests automatically that match
some pre-defined metrics and to trade
Some sites facilitate secondary trading of P2P loans: Users existing loan notes on dedicated
can trade existing loan notes on dedicated trading platforms. platforms. This injects new
This increases liquidity. New lenders can also diversify their professionalism into P2P lending but
portfolios across time by buying older loan notes.
also shifts power from people to
computers. back to front page
P2P sites in the US need approval from the Securities and
Exchange Commission .
Source: DB Research, 2009 6

November 2009 3
E-Banking Snapshot 31 Welcome to the machine

Borrowers need to explain their

Borrowers need to explain their situation situation. Borrowers at Lending Club
Length of loan description (no of characters), by credit grade use on average 370 characters to
500 describe their loan request. They write
450 more if their credit grade is bad. Each
400 notch downward (there are 35 from A1
350
300
to G5) is statistically associated with
250 3.5 additional characters. back to front
200 page
150
100
50
0
A B C D E F G

Credit grades decrease fom A (best) to G (worst)

Sources: DB Research, Lending Club, 2009 7

Ignore explanations at your own peril.

Mind their words Lengthy explanations tend to increase
Dependent variable: Repayment is late or in default (dummy) the chances of getting funded (not
Data sam ple: 10790 loans at Lending Club shown) but are also associated with a
Method: Logit regression higher delinquency risk – other things
being equal. As it appears, borrowers
Explaining variables with a more complicated background
Credit Quality 0.044*** need to explain that at length to
(One notch from good to bad) (0.006) prospective lenders who can then
Am ount borrow ed 0.090*** make informed choices. Lenders who
(USD '000) (0.007) ignore this information might load up
Length of loan description 0.213** too many bad loans during this
(in '000 characters) (0.100) downturn. back to front page

McFadden R2 20%

Levels of statistical significance: ***1%, **5%, *10%

Time trend accounted for. Sources: DB Research, Lending Club, 2009 8

© Copyright 2009. Deutsche Bank AG, DB Research, D-60262 Frankfurt am Main, Germany. All rights reserved. When quoting please cite “Deutsche Bank
Research”.
The above information does not constitute the provision of investment, legal or tax advice. Any views expressed reflect the current views of the author, which do
not necessarily correspond to the opinions of Deutsche Bank AG or its affiliates. Opinions expressed may change without notice. Opinions expressed may differ
from views set out in other documents, including research, published by Deutsche Bank. The above information is provided for informational purposes only and
without any obligation, whether contractual or otherwise. No warranty or representation is made as to the correctness, completeness and accuracy of the
information given or the assessments made.
In Germany this information is approved and/or communicated by Deutsche Bank AG Frankfurt, authorised by Bundesanstalt für Finanzdienstleistungsaufsicht.
In the United Kingdom this information is approved and/or communicated by Deutsche Bank AG London, a member of the London Stock Exchange regulated by
the Financial Services Authority for the conduct of investment business in the UK. This information is distributed in Hong Kong by Deutsche Bank AG, Hong
Kong Branch, in Korea by Deutsche Securities Korea Co. and in Singapore by Deutsche Bank AG, Singapore Branch. In Japan this information is approved
and/or distributed by Deutsche Securities Limited, Tokyo Branch. In Australia, retail clients should obtain a copy of a Product Disclosure Statement (PDS)
relating to any financial product referred to in this report and consider the PDS before making any decision about whether to acquire the product.

ISSN Internet: 1619-4829 / ISSN e-mail: 1619-6465

November 2009 4