Digital Agency Information Systems procurement Reform Study Meeting (2nd)

Overview

Date and time: Monday, August 1, 2022 from 13:00 to 14:30
Location: Online
Agenda:
1. Opening
2. Proceedings
  1. How to proceed after the second review meeting
  2. Summary of each discussion point
  3. Free discussion
3. Adjournment

Materials

Relevant policies

Ensuring fairness and transparency in procurement / procurement reform to utilize new technologies

Summary of proceedings

Date

From 13:00 to 14:30 on Monday, August 1, 2022

Location

Online

Members in attendance

Member Kajikawa, Member Arikawa, Member Kawazawa, Member Kimura, Member Sakashita, Member Sumiya

Summary of proceedings

Member Sumiya explained and reported on Issue B-2 "Development of the System procurement Platform," focusing on the digital marketplace, and each member had a free discussion. After that, the Secretariat explained A-3 "Ideal way of development and contract including agile procurement and quasi-mandate contract" and C-1 "Adoption of architecture to prevent vendor lock-in". After the explanation of each issue, the transition was held for free discussion and active opinions were expressed. The main opinions from each member are as follows.

B-2 "Development of System procurement Platform"

Committee Member Sumiya explained "Material 2: Overview of Digital Marketplace and Review Policy".
It is necessary to pay attention to the fact that the UK has a history of about 30 years, but the Japanese government plans to introduce the digital marketplace in one step. First, it is necessary to consider the case where the management method called P3RM adopted in the UK is applied to the Japanese government. In the case of the UK, 0.6% of the procurement amount is collected as a usage fee. It is also necessary to consider what the Japanese government will do.
As for the reduction of process costs, the cost inside the organization is higher than the external cost. The point is how much can be reduced, so I think it is necessary to actually try and make a validation.
I agree with the introduction of the digital marketplace, but I am worried about two points: the legal Issue and whether it can be used immediately at the site after introduction. Even if the legal Issue is solved in Digital Agency, it will take time to penetrate at the site. I think it will be used on a small scale at first, and will be gradually used in more and more cases.
I agree with the introduction of the digital marketplace, but depending on how it is advanced, it may not be used. How to realize it in the form in which each ministry actually participates is Issue. Although it has been pointed out for a long time, the separation and division ordering of information systems has not been realized, but I recognize that the possibility of realization is increasing due to changes in the situation. A local government that expands the use while accumulating results with each ministry, local design, and other public institutions is necessary. In addition, since the digital marketplace cannot cover all, it is necessary to explain what kind of procurement will be covered.
At present, in order to ensure competitiveness, the Administrative Reform Secretariat is investigating and trying to grasp how many contractors can receive orders in each contract ordering field, and recognizes that the consideration of the marketplace is an initiative in the same flow. If Issue can ensure fairness and clarity at the stage of registering in the marketplace first and at the stage of selecting and ordering from registered contractors second, I think that it can be introduced without contradiction with the current legal system.

A-3 Agile development, procurement Including Quasi-Mandate Contracts, and Contracts

As for the handling of discretionary contracts in foreign countries, the reason why the implementation conditions in the United Kingdom and the United States and the case of VFM (value for money) in Australia are mentioned is that discretionary contracts are irregular and have many problems, so careful check is required at both the entrance and exit of the contract. Therefore, it can be said that the expression "good cases of discretionary contracts" is a troubling expression. Discretionary contracts are also a form of contract permitted under the Accounting Act, and if the requirements are met, there is no problem in closing up as a reasonable case, but the expression "good cases" may be misleading.
It is necessary to create guidelines for contract changes, but basically, what is clearly stipulated in the Civil Code is interpreted and operated based on it even in the case of public-private contracts. Therefore, it may be difficult to include the content that is written in the Civil Code but not described in the Accounting Law only for contract changes. We would like you to create guidelines while keeping in mind that the main principles of the Civil Code, including default, are not described in the Accounting Law.
I understand that the Board of Audit pointed out that long-term renewal contracts could not be used in cloud service, but I believe that this is not the case. In the cases pointed out by the Board of Audit, the problem was that long-term renewal contracts for software and hardware were made together with telecommunications costs, and it is not that cloud service procurement could not be used.
In the case of a discretionary contract assuming an agile development, the point that performance by a person other than the contractor is "disadvantageous" may be a requirement. Therefore, I think that even under the current system, there is a standard to distinguish between a discretionary contract and a contract change that requires the same contract.
In relation to A-3 (a), even in the private sector contract, there is still a discussion about whether it is better to make the Agile development a contract. If it is promoted as a quasi-mandate contract as a policy, it is better to decide it. In addition, even though it is a quasi-mandate contract, there are cases where it is similar to a performance ratio type in which compensation is paid for work performed or a contract in which performance compensation is paid for certain deliverables, so it is necessary to specify what form will be adopted in Agile and create a model as a public authorities. In addition, in the context of cloud service, there was a talk about using a reseller as a way of using an intermediate business operator, but there is also a discussion that the use of a reseller will be recommissioned, so there is a concern that it will be necessary to adjust with the prohibition of recommissioning.
As for long-term renewal contracts, it is currently possible to conclude contracts only with water and gas, the Internet, and telecommunications businesses, but since there are many cloud service that are expanding as infrastructure in the real world, it is better to make it clear in terms of law that these can also be used.
Regarding examples of contract changes and discretionary contracts, it is sometimes observed that there is a lack of understanding of law in the first place, so it is considered useful to create guidelines.
As for the contract method, the United Kingdom has created a contract database and disclosed examples in it. It is helpful that the United Kingdom has concluded certificates of understanding between vendors and the government to give discounts. There is a sentence called the Budget Assessment Policy created by the Budget Bureau in Ministry of Finance in 2003. Since this year, the system has been changed so that it can be concluded for multiple years, so I think it is good to refer to the discussion when considering it.
We are also working on quasi-mandate contracts, and we have a policy of producing results while moving. To be specific, we do not include written specifications in the contract so as not to be a hard law, but a separate sheet. In addition, it is clearly stated in the contract that it is not a command and order because it is necessary to pay attention so as not to be a disguised contract. It is clearly stated that the ordering party will accept all intellectual property after delivery. As for the deadline, we have a policy of terminating the contract if the target is not achieved within three months to six months.
Regarding A-3 (a), I basically agree with the measures written by the Secretariat. Document 1 "How to proceed with the 2nd Review Meeting and the outline of each point" on page 19 says that the acceptance inspection "should be conducted at each sprint," but in order not to make Agile development itself heavy, shouldn't the acceptance inspection of each sprint be limited to the confirmation of important parts, and shouldn't consideration be given to the weighting, such as turning over the final part at the end? In addition, there are many policies for the creation of guidelines, but I felt that it is necessary to ensure that they are accessible by both the public and private sectors, and to create an environment in which the guidelines themselves can be updated by receiving feedback from users so that they do not become outdated. In addition, regarding training, training for product owners is important, but it is also important to obtain the understanding of people around them and create organization. For example, in Agile development, if people around them intervene or superiors request detailed reports, I think that product owners cannot exercise leadership and Agile development will not be possible. I think it is important to have people who are older than a certain age acquire knowledge through training, while having other members deepen their understanding.
We would like to confirm our basic understanding of the comments on long-term contracts. In the first place, the root of the idea of multi-year contracts is that the Fiscal Law clearly specifies four types of acts to incur national treasury debts, continuing expenses, permission carry-over, and accident carry-over and makes them exceptions, while there is a major principle of the Constitution that the people and the Diet, which is the representative of the people, should control the budget on a single-year basis. However, for those that bear the same amount of debts for the same amount of money every year, such as electricity and water charges, the Fiscal Law describes them as exceptions to exceptions based on the enumeration principle as long-term contracts. Around 2002, the Diet and the Board of Audit raised an issue because long-term contracts were used in a way to avoid acts to incur national treasury debts. If a long-term contract is used for system-related debt burden in the future, it will be problematic unless it is discussed in the Diet and specified in the law in the form of exceptions to the Constitution.
Long-term continuing contracts are listed in Article 102 (2) of the Cabinet Order, but when we look at examples in local government, there are cases where they are described more widely, so I think that is useful.
Regarding long-term contracts, I talked about the rules from the Constitution, but due to the exceptional treatment, a specific list must be stated. Whether it is really necessary or not should be validation, and whether it is up on the list or not should be considered, and the legal form must be based on the law. I would like you to carefully validation the content of the contract, and to consider stating it on the list of the government ordinance to the extent that it does not violate the law in accordance with the original rule system. In addition, it is a little dangerous to explore a third way, so I would like you to avoid it if possible.
The framework agreement method distinguishes between agreement and order placement, and it is possible to find a vendor who can receive an order in advance, so I think it will be beneficial to the competitiveness of procurement. I hope to consider and realize it as quickly as possible.

C-1 Adopting an Architecture That Prevents Vendor Lock-in

I think it is very important to utilize existing open source software. It is also important to open source the software itself entrusted to the government so that other administrations can use it, in the sense of eliminating vendor lock-in. In addition, in order to promote loose coupling of systems, it is very important to ensure interoperability. For example, the Five Digital Principles have been created in Digital Extraordinary Administrative Advisory Committee, but it is necessary to develop and penetrate a mechanism in which governments can move side by side, such as based on API disclosure and specific data standards, as stated in the Technology Code of Practice of the United Kingdom. In addition, in order to adapt the system to the outdated regulation, there are cases in which extremely complex systems are created, so it is important to change the Digital Agency in the opposite direction while cooperating with the regulation, and to set up a window where the outdated regulation can be shared. Regarding the discussion of intellectual property in C-1 (b), the Intellectual Property Rights Guide of the United Kingdom can be used as a reference, but in relation to the digital marketplace, I think it should be made clear so that administrative officials do not confuse the viewpoint that SaaS is different in the form of payment for licenses. In addition, regarding the previous point, I would like to emphasize that I would like the introduction of the framework agreement method to be proceeded with a sense of speed. development
Regarding vendor lock-in, the biggest problem is that there are no employees with IT knowledge who can write RFPs, and recruitment is being carried out with vague requirements and specifications that are advantageous to specific business operators. Therefore, the development of human resources is important. In the United Kingdom, there is a project management method called prince2, and seven themes appear in it, which are generally consistent with the architecture diagram of society5.0. I would like you to think about how to develop human resources by referring to such things.
It would be good if it could be open-sourced, but on the contrary, there are concerns that the number of people who can do it will decrease. The Bayh-Dor Act is difficult to use as a field sense. Although 100% of the rights are left to the business operator, it is necessary to report the status of use of the rights, it is necessary to obtain the consent of the government when sending it overseas, and it is necessary to grant the right to sublicense to the government. These are factors that make companies hesitate. I think the Bayh-Dor Act itself was active, but it is a little old, so it may be necessary to adjust the intellectual property part after changing it to a system that is easier to adopt the contract form of private sector.
In the Administrative Project Review, some people expressed concern about whether security is okay with cloud computing, and I think that open source is a high hurdle. Unless Digital Agency carefully explains to each ministry and department, and provides guidance such as open source for things that do not have problems in security, it will not spread as a whole-of-government initiative. In addition, data portability is very important. I feel that bidding is being carried out in a very difficult situation now, including the viewing of materials at the time of bidding consideration by business operators. At this point, from the viewpoint of data portability, we may proceed with individual consideration of various projects.
In the discussion of vendor lock-in, the Secretariat also felt that there might be concerns about the knowledge of the ordering party. I felt that the main premise of the discussion would be an awareness-raising campaign on how to enhance the knowledge of the ordering party in the future, and how to incorporate the knowledge and awareness of the entire organization, not just the department in charge of digital, into the administrative action itself. I would like to discuss the consultation counter, cooperation between ministries, etc. with the Secretariat.

End