Financial Innovation and Endogenous Growthhomes.chass.utoronto.ca/~xzhu/teaching/growth/Shiny.pdf · 2012-11-28 · Financial Innovation and Endogenous Growth Luc Laeven, Ross Levine,

Financial Innovation and Endogenous

GrowthLuc Laeven, Ross Levine, and Stelios Michalopoulos

NBER Working Paper No.15356

Summarized by: Shiny Zhang

Department of Economics, University of [email protected]

November 27, 2012

OutlineI Motivation

I Research Premises

I Contributions

I Model and its Predictions

I Empirical Evidence

MotivationI Historically, financial system improvements have been

integrally connected to technological change andeconomic growth

Historical Interactions between Technological and Financial Innovations

Time Location Technological

Innovation

Financial

Innovations

4000-3100 BC Sumerian-Uruk Labor

specialization

Tradable debt

contracts

800 BC Ancient Rome Large-scale mining

Limited liability

companies and

corporations

1870s-1900s United States Expansion of

railroads

Investment banks

and Credit bureaus

1959 United States Hi-Tech

Breakthroughs Venture capital

MotivationI Existing growth models assume:

I Financial system: exogenously given and inertI Technological innovation do not promote improvements

in financial system

Research PremisesI Goals:

I Model the endogenous and coevolution of finance andtechnology

I Provide emrpirical supports for the model predictions ofthe long-run interactions among, technology innovations,financial improvements and economic gowth.

Contributions - TheoreticallyI Textbook Schumpeterian models

I Profit-maximizing technological entrepreneursI Technological innovations over time

I Laeven, Levine and Michalopoulos (LLM)’s modelI Profit-maximizing financial entrepreneursI Financial innovations over timeI Model Prediction: The financial innovation rate speeds

up a country’s economic convergence toward thetechnological leader

I Contribute toward the building of a more general,dynamic theory of endogenous growth, financialinnovation, and financial regulation

The Basic Structure of the ModelI Begin with discrete-time Schumpeterian growth model

developed by Aghion, Howitt, and Mayer-Foulkes (2005)

I Assumptions - k countriesI no exchange goods or factors of productionI do adopt each others’ technological ideasI a continuum of individuals & a fixed population (N=1)

I Assumptions - individuals wih two-period life spanI utility function: U = c1 + βc2 where β ∈ (0, 1)I three types of labor specializations

I householdI entrepreneurI financier

The Basic Structure of the ModelI One final good - our numeraire

I To be used asI consumption goodsI input into entrepreneurial and financial innovationsI input for intermediate good production

I Production function

Zt = N1−α∫ 1

0A1−αi ,t xαi ,tdi ; α ∈ (0, 1), (1)

where Ai ,t is the level of technology, xi ,t is theintermediate good in sector i, and N is the labor supply.

I Perfectively competitiveness ⇒ price of eachintermediate good equals its marginal product

The Basic Structure of the ModelI A continuum of specialized intermediate goods

I In each sector i , a continuum of individuals with anentrepreneurial idea for period t is born in period t − 1

I Only one entrepreneur in a sector has a capable idea

Ai ,t =

{At with probability µei ,tAi ,t−1 with probability 1− µei ,t

}, (2)

where At is the world technology frontier and Ai ,t is thelevel of technology of intermediate goods sector i inperiod t.

I The quality of each entrepreneurial idea is unknownboth to the entrepreneur and to households.

I Households will only fund the entrepreneur designated ascapable by the financiers.

I The screening process for financiers is costly, risky, andpotentially profitable.

The Basic Structure of the ModelI Entrepreneurs or innovators

I Profits with price pi ,t = χ > 1

c =

{1 if successfulχ if unsuccessful

},

⇒ πi ,t =

{πAi ,t if successful0 if unsuccessful

}, (3)

where c is the units of final good as input needed perone unit of intermediate good production, and

π = (χ− 1)(αχ )1

1−α .I Successful innovators

I Earn monopoly profits for one period and die next periodI Technology can be imitated costless next period

The Basic Structure of the ModelI Financiers

I Each sector i has one person born each period t − 1 witha successful financial innovation that improves thescreening technology next period

I The level of screening technology in intermediate goodssector i in period t:

mi ,t =

{At with prob. µfi ,tmt−1 with prob. 1− µfi ,t

}, (4)

where At is the world technology frontier, and mt−1 isthe average of the screening technologies across allsectors in period t − 1.

The Basic Structure of the ModelI Financiers

I The probability that the financier in sector i correctlyidentified the capable entrepreneur:

λi ,t =

{At/At = 1 with prob. µf

i,t

mt−1/At = λt−1

1+g with prob. 1− µfi,t

}, (5)

where g is the growth rate of the world technologyfrontier, At is the world technology frontier, and mt−1 isthe average of the screening technologies across allsectors in period t − 1.

I Financiers are paid by entrepreneurs in the form of ashare, δi ,t , of entrepreneurial profits.

Timing of Events

Entrepreneurial InnovationI Capable entrepreneur screened by successful financiers

chooses µe∗i ,t to maximize profits

maxµei,t

Πei ,t = (1− δi ,t)(βµe

i ,tπAt − Nei ,t−1)

s.t. Nei ,t−1 = (θµe

i ,t)γAt , γ > 1,

(6)

where Nei ,t−1 is the amount of resources invested in

entrepreneurial innovation during period t − 1, θ is aneconomy-wide constant reflecting institutional and othercharacteristics.

Entrepreneurial InnovationI Imperfectly screened entrepreneur chooses µe′

i ,t tomaximize profits

maxµe′

i,t

Πe′

i ,t = βλi ,tµei ,tπAt − Ne

i ,t−1

s.t. Nei ,t−1 = (θµe

i ,t)γAt

(7)

where Nei ,t−1 is the amount of resources invested in

entrepreneurial innovation during period t − 1, θ is aneconomy-wide constant reflecting institutional and othercharacteristics.

Financial InnovationI The financier chooses to borrow and invest in financial

innovation to maximize profit:

maxµfi,t

Πfi ,t = βµf

i ,tδi ,tΠe∗i ,t − N f

i ,t−1

s.t. N fi ,t−1 = (θf µ

fi ,t)

γAt

(8)

where Nfi ,t−1 is the amount of resources invested in

financial innovation during period t − 1, θ is aneconomy-wide constant reflecting institutional and othercharacteristics.

Aggregating the Financial SystemI Average, or representative, probability that a financiers

successfully identified the capable entrepreneur

λt =

∫ 1

0

λi ,tdi

(9)

I Steady state level of average financial efficiency

λ∗ =µf ∗

g + µf ∗ (10)

I The higher is the steady state rate of financial innovation(µf ∗), the more efficient is the economy‘s financial systemat identifying capable entrepreneurs in the steady state(λ∗).

Equilibrium Economic Performance Across

CountriesI In the long run, the country’s technology gap

(at = At/At) converge to the steady state value:

ass =(1 + g)µ∗

g + µ∗ (11)

where µ∗ = µ(µf ∗, µe∗, λ)

Entrepreneurial Innovation

Lemma 1I Entrepreneurial innovation’s properties in sectors using

the old, imperfect screening technology:I The higher the profits (π) and the lower the cost of

innovation (θ), the more entrepreneurs invest ininnovation, which boost the probability of successfulinnovation:

∂µe′

i ,t

∂π> 0,

∂µe′

i ,t

∂θ< 0. (12)

I The rate of entrepreneurial innovation is an increasingfunction of the standard screening technology (λi ,t):

∂µe′

i ,t

∂λi ,t> 0. (13)

Aggregating the Financial System

Lemma 2I Steady State financial innovation’s properties

I The financial innovation is positively related to the

entrepreneurial innovation (∂µf ∗

∂µe∗ > 0)I The higher the costs of financial innovation (θf ), the

lower the Financial innovation (∂µf ∗

∂θf< 0).

I The high the innovation rate of the world technologyfrontier (g), the more probable the Financial innovation

(∂µf ∗

∂g > 0).

Major Model PredictionsI Proposition 1: An economy‘s steady state technology

distance (ass)’s properties:I Decreasing at the cost of financial innovation (θf )

∂ass∂µf ∗

∂µf ∗

∂θf< 0 (14)

I Increasing at the rate of entrepreneurial innovation (µe∗)

∂ass∂µe∗

∂µe∗

∂θ< 0,

∂ass∂µe∗

∂µe∗

∂π> 0 (15)

I Corollary 1: An economy blocking financial innovationwill eventually stagnate irrespective of the initial level ofscreening technology (λt)

ass = 0 if θf →∞ (16)

Model PredictionsI The model economy predicts that regardless of the

screening capability of the financial system in period t,anything that prohibits financial innovation will eventuallystop economic growth:

Empirical MeasuresI Main measure of financial innovation (f)

I the fraction of years a private credit registry existedwithin the priod 1960-95(Djankov et al., 2007)

I Main measure of financial development (F):I private credit to GDP in 1960 (Aghion, et al., 2005)

Measures - Growth and Productivity GrowthI Growth (g-g1)

I the growth rate of real per capita GDP of the countryminus the US growth rate in real per capita GDP(computed over the period 1960-95)

I Productivity Growth (pg-pg1)I average growth rate of multifactor productivity of the

country relative to the US computed over the period1960-95

Regression Frameworkd and Key ResultsI Regression Framework

g − g1 = b0 + b1F + b2(y − y1) + b3F (y − y1) + b4X

+b5f + b6f (y − y1) + u

I PredictionsI b1 and b5 will not be significantly different from zeroI b3 might be negative and significantI b6 will be negative and significant

Table 1: Financial Development, Financial

Innovation, and Growth(1) (2) (3) (4)

Variables g − g1 g − g1 g − g1 g − g1F −0.00498 0.00221 −0.0187

F ∗ (y − y1) −0.0299∗∗ −0.0268∗ −0.0336f −0.900 1.074 0.317

f ∗ (y − y1) −1.698∗ −1.880∗∗∗ −2.485∗∗∗

R − squared 0.558 0.585 0.409 0.242

Note: (1) is estimated using OLS and replicats AHM results; (2)is

estimated using OLS and adds screening innovation; (3)uses legal origin

and the change in financial reform index as instruments; (4) uses only te

legal origin of the country as an instrument. The symbols (*, **, ***)

denote significance at 10%, 5%, and 1%, respectively. g-g1 is the growth

rate of real per capita GDP of the country minus the US growth rate in

real per capita GDP (computed over the period 1960-95).

Table 2: Financial Development, Financial

Innovation, and Productivity Growth(1) (2) (3)

Variables pg − pg1 pg − pg1 pg − pg1F 0.00604 −0.0112

F ∗ (py − py1) −0.0172 −0.0464∗∗

f −0.608 0.498 −0.120f ∗ (py − py1) −1.900∗ −2.140∗ −2.540∗∗∗

R − squared 0.592 0.441 0.336

Note: (1) is estimated using OLS with screening innovation; (2) uses

legal origin and the change in financial reform index as instruments; (3)

uses only the legal origin of the country as an instrument. The symbols

(*, **, ***) denote significance at 10%, 5%, and 1%, respectively.

pg-pg1 is the average growth rate of multifactor productivity of the

country relative to the US computed over the period 1960-95.

Key Results and Policy ImplicationsI Key Results

I Economies without financial innovation stagnate,irrespective of the initial level of financial development

I A faster rate of financial innovation accelerates the rateat which an economy converges to the growth rate ofthe technological leader

I Policy ImplicationsI Adaptability and innovation are key elements for

sustaining economic growthI Institutions, laws, regulations, and policies that promote

financial innovation speed up the technological changeand economic growth

Table 1: Financial Development, Financial Innovation, and Growth

Table 2: Financial Development, Financial Innovation, and Productivity Growth

Measures - Instruments and control variablesI Instruments

I legal origin and the change (Abiad and Mody, 2005)

I Control variablesI same as in Aghion, Howitt, and Mayer-Foulkers (2005)

I education attainment (school)I government size (gov)I inflation (pi)I black market premium (bmp)I openness to trade (trade)I revolutions and coups (revc)I political assassinations (assass)I ethnic diversity (avelf)

I further robustness check: interations betweens thesecontrol variables, and y-y1, the log of per capita GDP in1960 relative to the United States

Table 3: Financial Development, Financial Innovation, and Growth: Other Interactions

Alternative MeasuresI Financial innovation (f):

I the fraction of years a public credit registry existedwithin the priod 1960-95

I the growth in the ratio of private credit to GDP over theperiod 1960-95

I Financial development (F):I the ratio of liquid liabilties to GDP in 1960I the ratio of bank assets to GDP in 1960I the index of creditor rights from La Porta et al. (1998)I the index of accounting qualiy in 1983 from La Porta et

al. (1998)

Table 4: Financial Development, Financial Innovation, and Growth: Other Proxies for

Financial Development

Table 5: Financial Development, Financial Innovation, and Growth: Other Proxies for

Financial Innovation

Extensions - Rent Seeking Financial InnovationI Capable financier investing in activities that increase the

production costs of the competitive fringe of financiers.

I Two-stage maximization problem for capable financiersI 1st: innovators choose screening-innovation investmentI 2nd: unsuccessful innovators choose rent-seeking

investment (b∗t ) to achieve the maximal expected profits(X (b∗t ))

maxbt

R(bt) = βbtΠe′t − Bt

s.t. Bt = (θbt)γAt

(17)

where bt is the fraction of the entrepreneurial profitsgiven to a rent-seeking financiers using the common,economy-wide screening methodology

Extensions - Rent Seeking Financial InnovationI Profit maximizaing rate of financial innovation

(µf ∗|rentseeking ) in the existence of rent-seekingopportunities

max Πft |rentseeking = βµf

t δtΠe∗t − N f

t−1 + (1− µft )X (b∗t )

Extensions - Lemma 3I The properties of financial innovation in the steady state

when rent seeking is possible:I Financial innovation under rent seeking is strictly lower

than financial innovation without rent seeking:

µf ∗|rentseeking < µf ∗ (18)

I Financial innovation increases as the cost of rent seekingincreases (θb)

∂µf ∗t∂θb

> 0 (19)

Additional Policy ImplicationsI Political, legal, and regulatory institutions that discourage

rent seeking and promote competition will tend to fostersocially beneficial financial innovations that facilitatetechnological progress.

I Institutions and policies that restrict competitioninfinance and provide a high return to rent-seekinginvestments by financiers will tend to retard financial andtechnological innovation.

I The LLM analysis also provides guidance both on how tomeasure financial innovation and on how not to measurefinancial innovation

Minor SuggestionsI Add the missing references by La Porta, Lopez-de-Silanes,

and Shleifer (1997, 1998)

I Explain in details how households allocate three units oflabor endowment in the first time period

ReferencesI Abiad A.and A. Mody. (2005). Financial Reform: What Shakes It?

What Shapes it? American Economic Review, 95, 66-88.

I Aghion, P., P. Howitt, and D. Mayer-Foulkes. (2005). The Effect ofFinancial Development on Convergence: Theory and Evidence,Quarterly Journal of Economics, 323-351.

I Djankov S., C. McLiesh, and A. Shleifer. (2007). Private Credit in129 Countries, Journal of Financial Economics, 84, 299-329.

I Laeven L., R. Levine, and S. Michalopoulos (2012). FinancialInnovation and Endogenous Growth, NBER Working PaperNo.15356.

I La Porta R., F. Lopez-de-Silanes, and A. Shleifer (1998). TheQualiy of Government, NBER Working Paper No.6727.

I La Porta R., F. Lopez-de-Silanes, and A. Shleifer (1997). LegalDeterminants of External Finance, NBER Working Paper No.5879.