Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Acquisitions, Productivity, and Profitability: Evidence from the Japanese Cotton Spinning Industry, Chad Syverson

Presentation by Chad Syverson, "Acquisitions, Productivity, and Profitability: Evidence from the Japanese Cotton Spinning Industry"

  • Be the first to comment

  • Be the first to like this

Acquisitions, Productivity, and Profitability: Evidence from the Japanese Cotton Spinning Industry, Chad Syverson

  1. 1. Acquisitions, Productivity, and Profitability: Evidence from the Japanese Cotton Spinning Industry Serguey Braguinsky, Carnegie Mellon University Atsushi Ohyama, Hokkaido University Tetsuji Okazaki, University of Tokyo Chad Syverson, University of Chicago
  2. 2. What We’re Looking At •The allocation of activity across production units matters to aggregate productivity •M&A can in principle reallocate control of productive assets to entities able to apply them more efficiently •Fits nicely with idea that managers and management practices shape productivity differences across plants, firms, and even countries
  3. 3. What We’re Looking At •But previous research has not been fully conclusive, partly due to several data problems –E.g., couldn’t cleanly distinguish between physical (quantity) productivity and revenue productivity –M&A may increase market power, leading to higher TFPR even if TFPQ is stagnant (or even falls) –Even if productivity increases, difficult to observe the mechanism
  4. 4. Why We’re Looking Here •Our data doesn’t have price-is-output problem while having lots of stuff one usually can’t observe (mechanism!) •We can measure, e.g.: –Profitability separately from productivity –Production process virtually at an engineering level –Productivity conditional on operation as well as capacity utilization (fraction of time in operation) –Firms’ product-market outcomes, connections, and inventory-holding behavior
  5. 5. Why We’re Looking Here •Our setting (Japan’s cotton spinning industry, 1890-1920) is also inherently interesting from a development point of view –Japan isolated until forced opening in 1860s –Cotton yarn and cloth accounted for fully one-third of imports during 1868-1875 –But soon became first world-class mfg. industry in Japan (Ohyama, Braguinsky, and Murphy, 2004) –Domestic production surpassed imports by 1890; exporting started 1895 and surpassed imports by 1897
  6. 6. Data •Decades of information on operations, management, and ownership of the universe of industry plants –Plant output and inputs in physical units –Output prices, inventories, and K utilization –Firm-level accounting data –Firms’ shareholder lists, board compositions, and business histories •Spans critical industrialization period for Japan
  7. 7. •After extensive growth through 1890s, consolidation –Better firms expanded extensively rather than intensively •73 acquisitions involving 95 plants during 1898-1920; 70% of industry capacity changed hands –Firms mostly joint stock companies with easily transferrable ownership (unusual for developing economies) –Rate of acquisitions higher than in comparable contemporaneous U.S. values What Role Did Asset Reallocation Play?
  8. 8. Number of Acquired Plants by Year
  9. 9. Main Findings (I) •More nuanced picture than the straightforward “higher productivity buys lower productivity” story •Acquired firms’ production facilities were not on average any less physically productive than plants of the acquiring firms, conditional on operating •But acquired firms much less profitable •This profitability gap did not reflect market power differences between the firms
  10. 10. Future Acquiring, Acquired and Exiting Plants before Acquisitions, 1896-97 Acquiring plants Acquired plants Exiting plants First cohort Second cohort Physical productivity (% diff. from average) 6.6 3.4 15.6 -21.1 Profitability (% return on equity) 27.4 18.5 15.9 15.9 Price (yen/400lb) 93.8 92.4 92.8 91.7 Days in operation per yr 324 316 301 279 Equipment age (years) 5.3 5.9 2.8 11.8 Number of plants 32 33 32 24
  11. 11. •Profitability—levels and changes—was not about prices (the usual suspect) •Most profitability differences before acquisition explained by input utilization •Acquired plants’ profitability growth came from increased utilization in short run and both utilization and productivity growth in the long run Main Findings (II)
  12. 12. Effects of Acquisitions, as Seen within Acquired Plants Acquired plants, relative to 3+ years before acquisition (% difference) Productivity ROCE Price 1-2 years before acquisition -0.3 (1.9) 2.0 (1.3) 1.1 (1.3) 1-3 years after acquisition 4.5* (2.6) 6.0* (2.2) 3.6 (2.7) 4-8 years after acquisition 12.6* (3.3) 8.9* (2.5) 4.4 (3.4) Observations 1,078 891 1,118
  13. 13. Acquired Plants Compared to Incumbent Plants of Their Acquirers Acquired plants, relative to incumbent plants of acquirers (% difference) Productivity ROCE Price Acquired plants, before acquisition -2.5 (2.1) -3.0* (1.1) -1.9 (1.4) Acquired plants, after acquisition 9.1* (2.3) 4.0* (1.4) 2.4 (1.7) Observations 1,487 1,392 1,528
  14. 14. Acquired Plants Compared to Incumbent Plants of Their Acquirers
  15. 15. Main Findings (III) •Profitability gap reflected systematically lower unit capital costs among acquirers –Higher capacity utilization (lower unit K costs) –Lower average inventory levels •These differences arise at least in part due to acquired firms’ deficits in demand management –Rationing occurred during times of low demand –Stronger companies had a broader network of connections at trading houses and dealt with more reputable buyers –New mechanism in the literature
  16. 16. Measuring Demand Management: Unrealized Output (I.e., Unsold Stuff) Share of unrealized output (%) Incumbent plants, 1-2 years before acquisition 3.3 Acquired plants, 1-2 years before acquisition 7.8 Acquired plants, 1-3 years after acquisition 3.2 Acquired plants, 4-8 years after acquisition 2.3
  17. 17. •We use 1898 nationwide registry to identify those most likely connected to cotton spinners’ output markets •Yielded list of 154 individuals –98 cotton yarn traders across Japan who paid more than 50,000 yen worth of operation tax that year –25 individuals listed as board members of the 4 largest incorporated cotton yarn-related trade companies –31 board members and traders registered at Osaka cotton and cotton yarn exchange •33 of 67 firms had at least one trader among board members and top shareholders—we call these “in-network” firms Firms’ Connections to Trading Houses
  18. 18. Firms’ Connections to Trading Houses Out-of-network In-network Physical productivity, % difference 0.0 5.5 ROCE, % 2.3 5.9 Unrealized output, % 12.7 8.4 Spindle utilization rates, % 73.9 78.1 Price, % difference from avg. -2.5 1.8 # of observations 127 170
  19. 19. Main Findings (IV) •Acquisitions thus best characterized as “higher profitability buys lower profitability” –Acquired plants weren’t less productive (when operating) because they had newer, better capital –However, this capital was being used suboptimally –New management raised both productivity and profitability •Leading firms set apart by better demand management and superior use of productive capital rather than market power •Acquisitions drove industry productivity growth
  20. 20. Conclusions •Asset turnover can happen for many reasons •Can be an important mechanism to raise productivity within an industry •This holds in our case, but mechanisms are subtle •Acquired firms not initially less productive, but rather less profitable –Their superior capital was underutilized •Acquisition by better managed and better connected firms increased both productivity and profitability of acquired assets

    Be the first to comment

    Login to see the comments

Presentation by Chad Syverson, "Acquisitions, Productivity, and Profitability: Evidence from the Japanese Cotton Spinning Industry"

Views

Total views

1,650

On Slideshare

0

From embeds

0

Number of embeds

1,059

Actions

Downloads

22

Shares

0

Comments

0

Likes

0

×