A comparative analysis of Innovation in addition to Productivity: lessons learned in addition to work ahead This is a joint ef as long as t with The challenge The difficulties Solutions

A comparative analysis of Innovation in addition to Productivity: lessons learned in addition to work ahead This is a joint ef as long as t with The challenge The difficulties Solutions www.phwiki.com

A comparative analysis of Innovation in addition to Productivity: lessons learned in addition to work ahead This is a joint ef as long as t with The challenge The difficulties Solutions

Davis, Roy, Contributing Editor has reference to this Academic Journal, PHwiki organized this Journal A comparative analysis of Innovation in addition to Productivity: lessons learned in addition to work ahead Chiara Criscuolo Centre as long as Economics Per as long as mance London School of Economics This is a joint ef as long as t with Australia: David (ABS) Austria: Martin Berger Belgium: Jeoffrey Malek Brazil: Bruno Araújo in addition to João De Negri Canada: Petr Hanel in addition to Pierre Therrien Denmark: Carter Bloch in addition to Ebbe Graversen Finl in addition to : Mariagrazia Squicciarini Olavi Lehtoranta Mervi Niemi France: Stephane Robin in addition to Jacques Mairesse Germany: Bettina Peters Italy: Francesco Crespi Mario Denni Rinaldo Evangelista in addition to Mario Pianta Japan: Tomohiro Ijichi (could not participate as long as data problems) Korea: Seok-Hyeon Kim Luxembourg: Anna-Leena Asikainen Netherl in addition to s: George van Leeuwen, Pierre Mohnen, Michael Polder, Wladimir Raymond New Zeal in addition to : Richard Fabling Norway: Svein Olav Nås in addition to Mark Knell Sweden: Hans Loof Switzerl in addition to : Spyros Arvanitis UK: Chiara Criscuolo A big THANK YOU to all! The challenge Comparative analysis of the innovation productivity link using a “structural” model Same model across countries: same variables; same estimation method So that estimates are “comparable” across countries Tools: Innovation Surveys Estimation routines

Pioneer Pacific College OR www.phwiki.com

This Particular University is Related to this Particular Journal

The difficulties Countries data cannot be pooled so each country must estimate model separately (in general no one outside country team can see the data) Not all countries could be at all meetings Innovation surveys similar (CIS “harmonized”) but not exactly the same across countries both in terms of variables used in addition to presence of “filter” questions HIGH COORDINATIONS COSTS!!!!!! Solutions Use the same model across countries: Minimum common denominator set of variables Use a model that control as long as selection as long as all countries “Centralize” implementation of estimation routines: Lead-country decides model Modify specific country model so that it is estimable in all countries (variables included – see also IPR topic; flexibility in using different variable names) Questions Was the centralized model the right one Should we have used a bottom up approach Should have more countries be involved with the as long as mulation of the model How useful was it to have the programming routine Should we change the approach as long as the follow-up project

A brief outline of the model followed To estimate the effects of innovation on productivity controlling as long as selection in addition to endogeneity Following the Crepon-Duguet Mairesse “tradition” we estimate a 3 stage/4 equation model: 1st innovation equation 2nd innovation input equation 3rd innovation output equation 4th productivity equation The model 3 stage with 4 equations 1st stage explains firms’ decision whether to engage in innovation activities or not in addition to the decision on the amount of innovation expenditure Prob(innovation=1)=f(size; Group; Foreign Market, Obstacles to innovation due to knowledge; costs in addition to market; industry dummies) Ln(innovation expenditure per employee)=f(Group; Foreign Market; Cooperation; Financial Support; industry dummies) In the 2nd stage we estimate the knowledge production function where innovative sales depends on investment in innovation. Ln(innovative sales per employee)=f(Innovation expenditure; Size; Group; process innovation; Cooperation with clients; suppliers; other private in addition to public agents; industry dummies; Mills ratio [to correct as long as selection]) The 3rd stage we estimate the innovation output productivity link using an augmented Cobb-Douglas production function using IV. Ln(sales per employee)=f(Size; Group; Process Innovation; log innovative sales per employee; industry dummies [Human Capital in addition to Physical Capital]) Obstacles to innovation

Cooperation partners Extensions/variations of the model Some countries could add Human Capital (H); Physical Capital (K) in addition to Materials (M) in the productivity equation: Austria (H,K); Belgium (H,K); Brazil (H,K,M); Canada (H,K); Finl in addition to (H,K,M); Germany (H,K,M); New Zeal in addition to (K,M); UK (H) Sales per employee is a very rough measure of productivity. Ideally we would want value added per employee (Labour Productivity measure) or Multi Factor Productivity measures To do this innovation surveys must be combined with other production datasets. (follow-up of current project) Extensions/variations of the model Most countries estimate separately as long as small vs large firms in addition to manufacturing vs services firms (excl. Italy in addition to Norway); Korea in addition to Canada only manuf; Luxembourg serv in addition to small. Important to look at differences St in addition to ard size threshold 250 employees but with some variation Issue as long as small countries in addition to /or small surveys problem of small sample sizes

Extensions/variations of the model Switzerl in addition to estimated a slightly different version of the model (e.g. no group variable; cooperation with as long as eign counterparts) Germany/Netherl in addition to s suggested a modification of the model to deal with endogeneity Canada: could only estimate on manufacturing in addition to weighted regressions in addition to no in as long as mation on obstacles to innovation Austria: estimate it on CIS3 rather than CIS4 Australia: no in as long as mation on as long as eign market; inputed group in as long as mation in addition to 2005 rather than 2004 New Zeal in addition to : again differences in variable definitions Extensions/variations of the model The original model only deals with selection in addition to endogeneity of innovation sales eq. In the productivity equation but we also wanted to deal with endogenity of innovation expenditure equation in the innovative sales equation: Ln(innov. sales per employee)=f(Inn. Exp.; Size; Group; process innovation; Cooperation with clients; suppliers; other private in addition to public agents; industry dummies; inverse Mills ratio) Option A: use predicted innovation expenditure rather than actual innovation expenditure in addition to bootstrap st in addition to ard errors Option B: instrument innovation expenditure in innovative sales equation (options suggested by Germany/Netherl in addition to s) Extensions/variations of the model In current version we use: “strict” definition of innovation: firms are innovative if innov. Exp.>0 in addition to innov. Sales>0 log innovative sales per employee Alternative version: “wide” definition of innovation: firms are innovative if innov. Exp.>0 in addition to innov. sales>=0 Share of innovative sales: use TOBIT rather than OLS as long as estimating innovative sales eq. Tested using data as long as Canada; UK; Denmark

RESULTS Simple “non-structural model” Probit Innovation OLS : innovation expenditure eq. OLS: productivity equation “structural” model: Heckman Innovative sales eq. Productivity eq. The results: Innovation equation The results: innovation investment equation

Results: the productivity equation Controlling as long as Selection: innovation equation (Heckman selection eq.) Heckman outcome equation: innovation expenditure eq. Careful: group in addition to as long as eign market are not marginal effects

Innovation Sales eq. Productivity equation Summary of results When significant, coefficients are surprisingly similar Serving a as long as eign market; being large in addition to being part of a group are generally associated with higher probability of being innovative in addition to financial support in addition to cooperation activity with higher investment in innovation Using a selection model is appropriate as long as most countries (exc. Austria; Luxembourg in addition to UK) In the innovation sales eq. the elasticity of innovative sales to innovation expenditure is mostly between 0.2 -0.35 In productivity eq. 0.3-0.6

Davis, Roy Computoredge Colorado Contributing Editor www.phwiki.com

Some counterintuitive results Obstacles to innovation have mostly positive coefficients. More innovative firms try harder in addition to there as long as e find more obstacles Process innovation is mostly negative in productivity equation. Measurement issues Adjustment costs (Possibly future work) Issues with the model Use of log innovative sales per employee Use of detailed cooperation variables in innovative sales eq. And cooperation in innovation exp. Eq. Use of process in both innovative sales eq. And productivity eq. Perhaps a simpler model E.g. only cooperation in innovative sales eq. And process in productivity eq. Lessons learned in addition to steps ahead What have we learned 18 countries! Interesting results High coordination costs/model more suited as long as some countries What next Match CIS with production data both as long as better productivity measure in addition to longitudinal dimension Alternative “organisational/coordination” model

Davis, Roy Contributing Editor

Davis, Roy is from United States and they belong to Computoredge Colorado and they are from  San Diego, United States got related to this Particular Journal. and Davis, Roy deal with the subjects like Computers

Journal Ratings by Pioneer Pacific College

This Particular Journal got reviewed and rated by Pioneer Pacific College and short form of this particular Institution is OR and gave this Journal an Excellent Rating.