Establishing a Framework for Managing Interest in Technical Debt

Areti Ampatzoglou, Apostolos Ampatzoglou, Paris Avgeriou

Department of Mathematics and Computer Science, University of Groningen, Netherlands

areti.ampatzoglou@rug.nl, a.ampatzoglou@rug.nl, paris@cs.rug.nl

Alexander Chatzigeorgiou

Department of Applied Informatics, University of Macedonia, Thessaloniki, Greece

achat@uom.gr

Keywords: Technical debt, Architecture, Software quality, Iinterest.

Abstract: Technical debt (TD) has gained significant attention over the past years. Due to its interdisciplinary nature, it

has become attractive for both technical and management stakeholders, to acknowledge and discuss issues

related to decayed design-time qualities over time, and their corresponding consequences. Until now, despite

the inherent relevance of technical debt management to economics, the TD research community has not suf-

ficiently exploited economical methods/models. Therefore, in this paper we present a framework for manag-

ing interest in technical debt, founded on top of well-known economic theories (i.e., Loanable Funds and

Liquidity Preference Theory) and current TD research. Specifically, in our framework, we will discuss aspects

related to technical debt interest, such as: types of TD interest, TD interest characteristics, and a proposed TD

interest theory. Finally, in order to boost the amount of empirical studies in TD research, we will propose

several tentative research designs that could be used for exploring the notion of interest in technical debt

practice.

1 INTRODUCTION

The term Technical Debt (TD) was coined in 1992 by

Ward Cunningham (1992) to describe the technical

compromises being made while coding, in order to

speed up product delivery and meet release deadlines.

Research on technical debt is rapidly growing over

the last years, since around 90% of articles on the sub-

ject have been published after 2010 (Li et al., 2015).

Similarly to its success among academics, TD seems

to be a topic that is appealing for practitioners, as

well. Specifically, according to Li et al. (2015), from

the current corpus of research efforts in technical

debt, 43% is performed in academia, 40% in industry

and 17% in both.

Apart from the fact that TD is a problem of para-

mount importance for software development, another

possible explanation for its popularity, in both aca-

demia and industry, is its interdisciplinary nature

(software engineering and economics), which facili-

tates the communication among technical and man-

agement stakeholders (Ampatzoglou et al., 2015). To

achieve this, the TD community borrows terms from

economics and maps them to software engineering

ones. Based on two recent literature reviews on the

subject (Ampatzoglou et al., 2015 and Li et al., 2015),

the two most frequently used financial terms in TD

research are: interest and principal.

Principal is a clearly defined concept, which is

characterized as the effort required to address the dif-

ference between the current and the optimal level of

design-time quality, in an immature software artefact

or the complete software system (Ampatzoglou et al.,

2015). Therefore, it is quantifiable and, in general, a

commonly accepted concept. On the other hand, in-

terest (associated with many definitions, which in

some cases are controversial) cannot be measured in

a straightforward way, since it involves the valuation

of future maintenance activities. Measuring interest

becomes even more complicated due to the fact that

its occurrence is not certain, in the sense that extra

cost/effort might not be required, and therefore inter-

est will not need to be paid off.

Additionally, research on TD interest and TD in

general, appears to lack empirical evidence. Accord-

ing to Li et al. (2015) 49% of the complete corpus of

TD research presents no empirical evidence, or only

toy examples, whereas this number rises to 56%,

Ampatzoglou A., Ampatzoglou A., Avgeriou P. and Chatzigeorgiou A.

Establishing a Framework for Managing Interest in Technical Debt.

DOI: 10.5220/0005885700750085

In Proceedings of the Fifth International Symposium on Business Modeling and Software Design (BMSD 2015), pages 75-85

ISBN: 978-989-758-111-3

when focusing on interest (Li et al., 2015).

To partially alleviate these problems, in this study

we investigate the notion of interest as it is applied in

the TD domain; our goal is to propose FItTeD, i.e., a

Framework for managing Interest in Technical Debt.

The FItTeD framework, aims to:

(G1) Identify types of TD interest, when it occurs,

and the high-level structure of its calculation.

Identifying the types of interest, which can oc-

cur along evolution, is the first step towards

more formal Technical Debt Management

(TDM). Until now, the definitions of interest

are rather high-level, and interest measurement

is often not applied in practice.

(G2) Explore how various characteristics of interest

in economics apply in TD interest. An example

of such a characteristic is whether interest is

simple or compound. However, these charac-

teristics have not been fully exploited in re-

search state-of-the-art, yet.

(G3) Propose a TD interest theory

. Until now, no

study has used the economic interest theories

for modelling technical debt interest. We will

rely on the Liquidity Preference Theory, for

modelling the evolution of TD.

The cornerstones for the development of FItTeD are:

 The corpus of existing research on Technical

Debt Management (TDM). We intend to reuse the

primary studies identified in a Systematic Litera-

ture Review (SLR) on technical debt by Am-

patzoglou et al. (2015), and filter them so as to ex-

tract primary studies related to interest, and syn-

thesize them in a systematic way (Kitchenham et

al., 2009).

 The existing economic interest theories. We in-

tend to apply existing economic interest theories,

i.e., the Loanable Funds and the Liquidity Prefer-

ence Theory, to reuse existing knowledge from

economics, on how interest should be handled,

and learn from accumulated experiences.

This framework aims at supporting software engi-

neers to determine the change of technical debt

amount in the future, by holistically describing all pa-

rameters that affect its future value (i.e., repayment,

interest, additional debt, etc.). This can in turn allow

the use of elaborate financial methods in several tech-

nical debt management activities, i.e., repayment,

monitoring, and prioritization. Additionally, we ex-

pect that the proposed framework can boost empirical

research in the field of TD, in the sense that it can

facilitate a common understanding on TD interest and

point to interesting research directions.

The rest of the paper is organized as follows: In

Section 2, we present related work from the field of

economics, i.e., the dominant interest theories. Next,

in Section 3, we will present the outcome of revisiting

the primary studies of the SLR by Ampatzoglou et al.

(2015), by presenting only interest-related infor-

mation. In Section 4, we present the proposed frame-

work for managing interest in technical debt. In Sec-

tion 5, we discuss possible ways that our framework

can be used for boosting empirical research in the

field of TD. Finally, in Sections 6 and 7, threats to

validity and conclusions are presented.

2 INTEREST IN ECONOMICS

Regarding the way interest rate is defined in the mar-

ket; various models have been suggested, by different

schools of economics (Mishkin and Eakins, 2012).

The mainstream theories are the Loanable Funds The-

ory, developed by the neoclassical school, and the Li-

quidity Preference Theory, proposed by the Keynes-

ian theory (Mishkin and Eakins, 2012).

Interest rate is the price paid for borrowing money

or vice versa (the payment received to loan money).

Therefore it can be considered as the price of money.

Interest rate, as any other price, can be defined in the

market at the equilibrium between supply and de-

mand. According to the Loanable Funds Theory, in-

terest rate specification takes place in the market of

loanable funds. On the one hand, individuals or enter-

prises, who want to invest, form the demand for loan-

able funds. They ask for loans in order to proceed

with an investment. As interest rate gets higher, bor-

rowing becomes more expensive. As a result, demand

for loanable funds decreases as interest rate increases.

On the other hand, the supply of loanable funds

comes from people or enterprises that use the loana-

ble funds market to save their money. Instead of con-

suming part of their income, they choose to put it into

the loanable funds market in order to save it for later.

In this case, higher interest rate means higher return

on savings. Therefore, supply of loanable funds rises

as interest rate increases.

In the diagram of Figure 1, the equilibrium in

loanable funds market is presented. We note that, in

economic theory, all kinds of supply – demand dia-

grams represent the dependent variable on the hori-

zontal axis and the independent variable on the verti-

cal axis. Therefore, in this case, the vertical axis de-

picts interest rate (r), while the horizontal axis repre-

sents the quantities of supply and demand for loana-

ble funds. The quantity of loanable funds supplied at

any level of interest rate is presented by line S. Line

Fifth International Symposium on Business Modeling and Software Design

S depicts the positive correlation between interest rate

and loanable funds supply. Likewise, the quantity of

loanable funds demanded at any level of interest rate

is presented by line I. The negative correlation be-

tween interest rate and loanable funds demand is in-

dicated by the negative slope of line I. When interest

rate is higher than r*, then it is more profitable to

save, or it is more profitable to lend than to borrow,

and supply of loanable funds is higher than demand.

On the other hand, when interest rate r is lower than

the level of r*, then it is more profitable to invest, or

it is more profitable to borrow than to lend, and de-

mand for loanable funds is higher than supply. When

r=r*, then both the investors and the savers have no

motivation to change their position in the market and

equilibrium is achieved. Consequently, interest rate is

determined at r=r*.

Figure 1: Loanable Funds Theory.

Equilibrium in the market is achieved at interest

rate r*, when every other factor, that could influence

savings or investment, is considered stable (ceteris

paribus – i.e., a Latin phrase, often used in economics

to suggest that all other factors are constant, in order

to examine the relationship between two variables).

Therefore, interest rate level may move upwards or

downwards in case of changes to savings or invest-

ments, due to exogenous factors (e.g., income). For

example, an increase in income would cause an in-

crease in the quantity of savings. That would result in

a shift to the right of the savings curve (S), which is

the supply of loanable funds. In Figure 1, the new line

depicts such a change. As shown in the diagram,

the new equilibrium is now achieved at point E

and

interest rate is defined at r

, lower than r*.

The Liquidity Preference Theory determines in-

terest rate level through the mechanism of supply and

demand for money (cash), which is performed in the

money market. In this case, supply of money (M) is

given at any point of time and is determined by the

central bank, according to the needs of the economy.

In other words, supply of money is not dependent on

interest rate and it is exogenously defined. On the

other side, demand for money (L) represents the

quantity of cash that people prefer to hold for pur-

poses of transactions, precaution or speculation. In

this case, as interest rate increases, it becomes more

profitable for people to invest money than to hold it.

Consequently, an increase in interest rate leads to a

decrease in the quantity of money demanded in the

market and a decrease in interest rate causes an in-

crease in demand for money. Similarly to the Loana-

ble Funds theory, interest rate is determined by the

equilibrium point of the market.

Figure 2: Liquidity Preference Theory.

The diagram of Figure 2 shows the equilibrium in

the market of money. Interest rate is represented on

the vertical axis, whereas money supply and demand

are shown on the horizontal axis. The supply curve is

vertical to the horizontal axis, and represents the sta-

ble money supply, provided by the central bank, in-

dependently of the interest rate level, as mentioned

above (this assumption consists the main difference

with the loanable funds theory). Demand for money

is negatively related to interest rate (because in this

case interest rate is the cost of holding money against

to investing in a bond) and line L shows the quantity

of money demanded at any given interest rate, ceteris

paribus. The intersection of the two curves, M and L,

represents market equilibrium and determines the

level of the interest rate at r*.

In case of a change in demand for money because

of a change in another determining factor, e.g. in-

come, or in case of a change in the quantity of money

supplied by the central bank, equilibrium rate will

change. For example, if the central bank decides to

increase money supply, then M would increase to M

and the curve in the diagram of Figure 2 would shift

Establishing a Framework for Managing Interest in Technical Debt

to the right. Consequently, equilibrium would be de-

fined by point E

and the new interest rate in the mar-

ket would be r

, lower than r*.

3 INTEREST IN TECHNICAL

DEBT RESEARCH

In this section we present an overview of studies that

have investigated interest in Technical Debt Manage-

ment (TDM). According to Ampatzoglou et al. (2015)

and Li et al. (2015), interest is the prominent financial

term that is used in TDM research. Note that in eco-

nomics, interest theories are used for calculating in-

terest rate (not interest per se), since interest is calcu-

lated based on interest rate. However, in TDM inter-

est is not calculated based on interest rate, but it is

assessed in various other ways, as explained later in

this section. Specifically, from TD research, it is not

clear if interest rate can be defined at all. In this study,

we reuse the dataset extracted by Ampatzoglou et al.

(2015), i.e., 29 studies that focus on TD interest. In

this paper, we are not presenting in detail the SLR

process, since it is thoroughly discussed in the origi-

nal study, but only an outline:

 Queried 7 digital libraries (IEEE, ACM, Scopus,

Springer, Science Direct, Web of Science, and

Google Scholar), with the term technical debt.

The search returned 1,173 primary studies

 Applied Inclusion/Exclusion Criteria (e.g., is the

study focused on the financial aspect of TD). The

process returned 69 primary studies.

From that stage and on, the process is specialized

to the goals of this paper. Specifically, first we filtered

primary studies related to interest. This step has been

performed as part of data collection in the original

SLR. Therefore, in this study we explored the 29 pri-

mary studies, which according to Ampatzoglou et al.

(2015) are relevant to interest. This set of studies is

our primary study dataset. For each study, the follow-

ing data have been extracted:

] Interest amount definition. We record the defi-

nition that the authors provide for the amount of

interest. The term interest amount is derived by

the work of Seaman and Guo (2011), who sug-

gest that interest should be calculated by taking

into account two components interest amount

and interest probability (see D2).

] Interest probability definition. We record how

interest probability is defined and calculated.

] Evolution of Interest. We record any possible

discussion that is related to how TD interest

amount grows or shrinks, along evolution. For

example, we capture if a study characterizes in-

terest as compound or simple, or as continuously

increasing.

] Interest estimation method. We describe how

TD interest is quantified in the primary study

(when applicable).

The mapping between data extracted and the goals set

in Section 1, are discussed below:

G1: We use [D

] and [D

]. Based on the frequency of

each variable, we extract the most common def-

initions of interest amount and interest probabil-

ity.

G2: We use [D

] that is related to studies, which dis-

cuss the evolution of technical debt interest.

Based on existing literature, and the definitions

derived from G1, we formulate the evolution of

TD interest, and investigate cases when it is in-

creasing or decreasing.

G3: We use [D

] that aims at describing how each

study assesses the amount of interest or the in-

terest probability, and synthesize them with the

financial interest theories and the definitions de-

rived from G1, to develop an interest theory that

is applicable for TD.

The outcome of the data collection phase is presented

in Table 1 and Table 2. Specifically, in Table 1, we

present data D

and D

; whereas in Table 2, we pre-

sent data D

and D

. We note that due to space limi-

tations: (a) in both tables, the citation is provided with

limited identifiers needed for characterizing a study

(e.g., omitting “et al.”), and (b) in Table 2, we only

present studies that hold a value for at least one vari-

able.

From Table 1, we can observe that about 31% of

primary studies describe interest amount as the extra

effort during maintenance, whereas 51% as the extra

maintenance cost. However, since in software econom-

ics cost is usually defined as a function of effort, we

can assume that 82% of studies refer to interest amount

as the extra effort/cost that is evident during mainte-

nance activities, due to the presence of technical debt.

The rest of the studies, either provide more high-level

definitions – i.e., (Eisenberg, 2012) and (Letouzey,

2012) – or define technical debt interest, similarly to

economics, i.e., the increase rate of technical debt

amount (Ernst, 2012), or define interest as a change in

a design-time quality attribute – see for example (Sea-

man et al., 2012) and (Zazworka et al., 2011). Addi-

tionally, we can observe that approximately 28% of the

studies acknowledge the existence of interest probabil-

ity. From these studies, two – i.e., (Guo and Seaman,

2012) and (Snipes, 2012) – adopt a financial risk man-

agement approach where interest probability is calcu-

lated as the standard deviation of interest rate; whereas

Fifth International Symposium on Business Modeling and Software Design

the rest adopt a risk management approach, i.e., they

consider interest probability as the probability of the

TD incurring event to occur.

Table 1: Data Extraction Overview. (1/2).

Study

Interest Amount

Interest Probability

Allman (2012) Increased effort to maintain and extend the

system

Alzaghoul (2013) Cost incurred by time due to an investment at

service level which is not properly managed

Brown (2010) Increased future costs owing to earlier quick

and dirty design and implementation choices

The probability that a particular type

of TD will have visible consequences

Buschman (2011) Cost to be paid later due to quick develop-

ment

Chin (2010) Cost of organization to hold on TD, plus the

additionally incurred debt

Codabux (2013) Additional cost of not eliminating TD now

Curtis (2012,

Software)

Continuing costs attributable to should-fix vi-

olations that haven't been remediated

Curtis (2012,

MTD)

Continuing costs attributable to should-fix vi-

olations that haven't been remediated

Eisenberg (2012) Long-term impact of TD

Ernst (2012) The rate of increase in TD

Falessi (2013) The cost that will occur by not fixing the

technical problem

Interest is not certain. It has a proba-

bility to occur, changing over time

de Groot (2012) The difference in cost between maintenance

at the ideal level and any level below

Guo and Seaman

(2011)

Extra work that will be needed if TD item is

not repaid

Interest standard deviation, because

of the uncertainty of interest

Guo et al. (2011) Additional cost

Holvitie (2013) The amount of extra work the principal can

cause to future development

The probability of extra work TD can

cause to future development

Koolmanojwong

(2013)

More expensive to fix than it is to do it right

the first time

Letouzey (2012) The negative impact of TD

Marinescu (2012) Extra maintenance effort required in the fu-

ture due to hasty, inappropriate design

McGregor (2012) Any extra work over the expected amount,

when later we carry out the deferred activity

Nord (2012) Increasing rework cost of the unpaid TD

Nugroho (2011) The extra maintenance cost spent for not

achieving the ideal quality level

Schmid (2013) Additional effort spent on not quite good code

Seaman (2011) Potential penalty paid in the future as a result

of not completing tasks in the present

The probability that TD, if not repaid,

will make other work more expensive

Seaman (2012) Decreasing maintainability The probability that TD, if not repaid,

will make other work more expensive

Siebra (2012) Extra Effort

Snipes (2012) The extra cost required to complete a mainte-

nance activity in the future if the task is post-

poned, plus the cost of other work that is re-

quired due to the presence of the TD

Interest standard deviation, because

of the uncertainty of interest

Establishing a Framework for Managing Interest in Technical Debt

Study

Interest Amount

Interest Probability

Zazworka (2011) Impact on quality

Zazworka (2013) An estimate of the amount of extra work that

will be needed if this TD item is not repaid

The probability that TD, if not repaid,

will make other work more expensive

Zazworka (2014) Probable future cost of not fixing the TD

Table 2: Data Extraction Overview. (2/2).

Study

Interest

Evolution

Estimation Method

Allman (2012) Compound -

Buschman (2011) Compound -

Chin (2010) Both -

Codabux (2013) Increasing -

Guo and Seaman (2011) Expected interest amount and interest standard deviation can be esti-

mated using historical effort, usage, change, and defect data.

Guo et al. (2011) Interest = interest amount × interest probability

IA = X – P,

X: Cost of doing something at t

(after postponing at t

), P: principal

Nord (2012) Increasing -

Nugroho (2011) interest would be the difference between maintenance effort spent at the

5-star level and any of the lower quality levels

ME = MF*RV/QF

MF=Maintenance Fraction (Historical Data), QF=Quality Factor,

RV=Rebuild Value (estimate of effort to be spent to rebuild a system)

Seaman (2011) Interest amount = W × C, C=average cost of the last N modifications to

module, W=weighting factor , based on the initial rough estimate (high,

medium, or low) of the interest amount

Siebra (2012) Increasing Estimation based on documentation (chronograms, backlogs and code

lines modifications) as the total effort between alternative scenarios

Furthermore, the results of Table 2, suggest that ap-

proximately 21% of primary studies deal with the evo-

lution of interest along time and either characterize it

as compound, or continuously increasing. As an excep-

tion to this, Chin et al. (2010), proposes that one type

of interest is simple. Specifically, they suggest that the

cost of the organization to hold on TD is stable across

time and neither increases nor decreases.

Finally, only 17% of studies propose a specific way

of measuring interest. The estimation is in most of the

cases performed by using historical data, documenta-

tion, and maintenance effort estimation models (for de-

tails see Table 2).

4 FRAMEWORK FOR

MANAGING INTEREST IN TD

In this section we present FItTeD, i.e., the proposed

framework for managing interest in technical debt.

While presenting FItTeD, the discussion focuses on

goals G1 – G3, as set in Section 1. The proposed

framework is based on the findings discussed in Sec-

tion 3 and on the general perception of interest as the

extra effort required for performing any maintenance

tasks when technical debt has been accumulated.

However, it has been enhanced, by our own sugges-

tions to cover gaps in the current literature.

Fifth International Symposium on Business Modeling and Software Design

4.1 Types of Interest

From the technical debt literature it is evident that

technical debt interest is perceived as a risk for soft-

ware development, in the sense that it has a specific

effect (i.e., interest amount) and a probability to occur

(i.e., interest probability). Concerning the amount of

interest, we assume that interest can be accumulated

through the extra cost incurred by two activities:

 Interest while repaying TD – I(r): The effort for

repaying technical debt at any time point t (i.e.,

enhancing the quality of a Technical Debt Item -

TDI) is higher than the effort needed for repaying

technical debt for this item, at any time point prior

to t. Therefore, I(r) is calculated as the differ-

ence between the two aforementioned efforts.

This type of interest will occur when (and if) the

amount of TD is to be paid off.

 Interest while performing maintenance activities

– I(m): Performing maintenance tasks is more

time/effort consuming in parts of the software

with accumulated TD, compared to parts in which

TD is reduced or zero. The difference between the

two amounts of effort is the amount of the I(m)

interest. This type of interest will occur, and will

be simultaneously repaid, when maintenance

tasks are performed (i.e., while undertaking the ef-

fort to perform the maintenance task).

Both the aforementioned types of interest are in

agreement with the most established definitions of in-

terest amount (i.e., extra cost/effort); however by add-

ing more details on when these extra costs/efforts can

occur. Thus, for each technical debt item, interest

TDI

) should be calculated, based on the following

high-level formula:





=

(



)

+

(



)

=

(



)

∗

(



)

+

(



)

∗

(



)

in which P denotes the probability of a repayment or

maintenance event to occur, E the effort needed to

perform an action, r denotes repayment, and m other

maintenance activities. To transform the aforemen-

tioned formula from the TDI level to the system-level,

we propose the use of the sum aggregation function,

in the sense that the total TD of a system is the sum

of TD, of all items with incurred TD. Therefore, in-

terest at system level (I) can be calculated, as fol-

lows:

=  



∗



+



∗





()



We note that the aforementioned formulas cannot

be used per se, but should be instantiated from re-

searchers, by conducting empirical research that

would assign estimates for the P and E factors. For

examples and interesting research directions on this

issue, see Section 5.

4.2 Evolution of Interest

Based on economics, interest is classified over two

dimensions: its method of calculation and its variation

over time. For these purposes, interest can be:

 Simple or Compound: Interest is simple when it

is calculated only as a function of the principal;

whereas it is compound when it is calculated over

the principal, plus the incurred interest; and

 Fixed or Floating: Interest rate is fixed, if it does

not change along time; whereas it is floating when

it can increase or decrease based on circum-

stances.

Technical debt literature has discussed these charac-

teristics of interest, but only superficially, without em-

pirical evidence on the real-world evolution of inter-

est. As already explained in Section 1, interest rate is

not defined in technical debt. Therefore, the distinc-

tion between floating and fixed interest rates is not ap-

plicable. However, interest amount can still increase

or decrease, based on the amount of debt that it is cal-

culated upon. To this end, we note that studies which

refer to continuously increasing interest are referring

to debt amount and not interest amount.

From observing the literature, we can claim that

researchers perceive technical debt interest as com-

pound, in the sense that it is increasing, since the ad-

ditional effort to repay technical debt and perform

maintenance on a technical debt item increases as soft-

ware grows. At any specific point in time (t

), it is

non-trivial to decompose the complexity of the system

to the original system complexity (C

), i.e., the one

that existed in the system when the principal incurred,

and the additional system complexity (C

), i.e., the

one that incurred due to system evolution (system

larger in size, more functionality, etc.). Therefore, the

calculation of the effort needed to perform any

maintenance action in t

, can only be assessed based

on system current complexity (C

However, interest is not expected to be continu-

ously increasing. We expect that such a claim only

holds for cases when no repayment activities are per-

formed. Specifically, in case that some repayment ac-

tivity is performed (at t

), we expect system complex-

ity after partial repayment (C

) to decrease (i.e., C

), leading to a decreased amount of both types of

interest, in future maintenance activities - E(r|m).

Establishing a Framework for Managing Interest in Technical Debt

These claims are valid for individual TDIs, in which

no additional technical debt has been incurred be-

tween timestamps t

and t

; and

summarized as fol-

lows:

Evolution















(

|



)



(

|



)

,

(





)

=0



(

|



)



(

|



)

,

(





)







(

|



)



(

|



)

,

(





)







For example (2

clause): in case the effort spent at

time point t

to partially repay technical debt E(r

)

is lower than the additional interest incurred at t

then it is reasonable to assume that any future mainte-

nance or repayment effort E(r|m

) will be higher

than the corresponding effort required at t

E(r|m

in the sense that the amount of debt (diminished de-

sign-time quality or complexity) is larger at t

com-

pared to t

4.3 Interest Theory

Based on the above, and by borrowing the rationale

of the equilibrium achievement from the existing eco-

nomic interest theories, we have been able to develop

an interest theory for managing TD interest. Specifi-

cally, we adopt the concept of the Liquidity Prefer-

ence Theory. The reason for selecting the Liquidity

Preference Theory and not the Loanable Funds The-

ory is that in TD the amount of money that is available

to the company for managing technical debt is stable,

i.e., the amount that has been saved, while incurring

TD – i.e., the principal (supposing that principal is not

invested, to provide extra benefits). The assumption

that the available money for managing TD is princi-

pal, is based on the fact that principal is the maximum

amount that can be spent without spending any addi-

tional effort (other than the one saved).

In the proposed interest theory, we map money

supply to principal, in the sense that principal is the

amount of money that is available to the software de-

velopment company, after incurring TD; and the

money demand to the accumulated amount of inter-

est, in the sense that this is the extra amount of money

that is demanded by the company when perform fu-

ture maintenance activities, caused by the TD. In Fig-

ure 3, where we present the FItTeD Interest Theory,

the x-axis represents time, whereas the y-axis repre-

sents amount of money. Therefore, the equilibrium

point (E

) denotes the time stamp (t

), in which the

company has spent the complete amount of money

from the internal loan (i.e., initial principal – P

) in

extra maintenance activities because of the incurred

TD.

We note that the specification of the equilibrium

point is achieved through an analysis based only on

effort, i.e., the effort saved when taking on TD and

the extra effort required for any future maintenance

activity because of its accumulation. Any other re-

lated costs or benefits related to technical debt occur-

rence (e.g. gains from launching the product earlier)

have been excluded from the model for simplicity

reasons. Thus, if the expected lifespan of the specific

TDI is shorter than t

then undertaking technical debt

is a beneficial choice, whereas if not, technical debt

becomes harmful for the company. The aforemen-

tioned discussions, in the case that no repayment ac-

tions are performed, are summarized in the blue lines

of Figure 3.

Figure 3: FItTeD Interest Theory.

Additionally, in Figure 3, we consider Σ(Ιm) as

continuously increasing, since it is a sum of positive

numbers and as exponentially increasing, because TD

interest is compound (see Section 4.2). In case that

some repayment occurs at some timestamp (t

), the

line of the accumulated interest Σ(Ιm)is moved up-

wards, due to the interest paid for repayment – i.e.,

I(r

) – but its slope is decreasing, since the interest

is expected to lower for future maintenance activities

(Im). This in turn leads to a shift of the equilibrium

point (E’) to the right, increasing the benefit period

’). The fact that principal is lowered to P

), is not presented in the diagram since the money

supply line (P

) is not moved, because the originally

available budget of the company is not affected. The

proposed interest theory can help practitioners in their

decision making by:

 Identifying the timestamp in which incurring TD,

becomes harmful for the company. Thus, they can

decide if they should undertake the debt.

Fifth International Symposium on Business Modeling and Software Design

 Supporting them on continuously monitoring the

interest that they have paid so far.

 Evaluating the repayment activity, based on the

time-shift of the equilibrium point that it offers.

5 RESEARCH IMPLICATIONS

As already discussed in Section 3 research on TD in-

terest is very theoretical and lacks empirical evidence.

Therefore, in this paper we aim at pointing out spe-

cific research directions, which would boost the em-

pirical research related to TD. The results of these

empirical studies would provide data for the instanti-

ation of the FItTeD interest theory. We organize the

tentative research design by goal:

Types of Interest: An interesting research direction

could be the empirical investigation of:

 whether I(r) and I(m) occur with the same fre-

quency, and

 whether I(r) and I(m) produce a similar amount

of interest when they occur,

 how I(r) and I(m) amount could be modelled, as

a function of the principal, or the underlying struc-

ture of the TDI.

So far, these questions have been explored only by

Guo et al. (2011), Nugroho et al. (2011), and Siebra

et al. (2012), by exploring historical changes and doc-

umentation. The research state-of-the-art lacks real-

world evidence on effort allocation.

Evolution of Interest: A possible empirical investiga-

tion of the evolution of TD interest could reveal inter-

esting characteristics of TD, such as:

 What is the relationship of the decay of quality in

the underlying system structure and the increase

in E(m) or E(r)? Answering this question could

guide practitioners on how to model the increase

of interest during software evolution.

 How frequently is E(r

) higher or lower than

I(r

)? Answering this question could unveil the

frequency with which repayment activities can

constitute interest increasing or decreasing.

FItTeD Interest Theory: In order to increase the ap-

plicability of the proposed TD interest theory, the fol-

lowing questions need to be empirically explored:

 What is the average time-shift that is benefited

from performing specific repayment activities?

 From what factors is this time-shift influenced?

 What is the relationship between I(r) and the av-

erage decrease in the I(m) of future maintenance

activities?

Answering these questions, would enable practition-

ers to instantiate the proposed interest theory, based

on real and context-specific data, and transform FIt-

TeD into a useful tools for practitioners.

6 THREATS TO VALIDITY

In this study, we actually inherit all threats to validity

from the original SLR on which we have based our

results upon (Ampatzoglou et al., 2015):

 the identification of primary studies

 the generalization of results, and

 the conclusions

Concerning data extraction, since we inde-

pendently performed this step, the corresponding

threats are related only to this study. To mitigate bias,

while extracting data, two researchers performed data

collection independently, compared the results and

discussed possible differences. The final dataset was

built through the consent of all authors. Finally, as a

threat we acknowledge that the construction of the

presented formulas, is to some extent based on the un-

derstanding of the authors on TD interest.

7 CONCLUSIONS

Nowadays, Technical Debt (TD) is receiving increas-

ing interest by both academia and practitioners, lead-

ing to an explosion of studies in this field. The cor-

nerstones of TD are two notions borrowed from eco-

nomics: i.e., principal and interest. Although princi-

pal is a well-established term, interest has so far been

discussed in a rather coarse-grained way, with several

contradictions among researchers.

In this paper, we propose FItTeD, i.e., a frame-

work for managing interest in TD, which takes into

account existing TD literature and economic interest

theories. The framework comprise of: (a) a TD inter-

est definition, (b) a classification of TD interest types,

a TD interest theory, based on the Liquidity Prefer-

ence Theory. The proposed framework is expected to

aid in the decision making of practitioners, and points

to interesting research directions. The main emphasis

of the future research directions is on empirical stud-

ies, which until now are underrepresented in the TD

research corpus.

REFERENCES

Allman E., 2012. Managing technical debt. Communica-

tion, 55 (5), pp. 50-55, ACM.

Establishing a Framework for Managing Interest in Technical Debt

Alzaghoul E., Bahsoon R., 2013. CloudMTD: Using real

options to manage technical debt in cloud-based service

selection. 4

International Workshop on Managing

Technical Debt (MTD ‘13). IEEE Computer Society.

Ampatzoglou A., Ampatzoglou A., Chatzigeorgiou A., Av-

geriou P., 2015. The Financial Aspect of Managing

Technical Debt: A Systematic Literature Review. Infor-

mation and Software Technology, 64, pp. 52-73, Else-

vier.

Brown N., Cai Y., Guo Y., Kazman R., Kim M., Kruchten

P., Lim E., McCormack A., Nord R., Ozkaya I.,

Sangwan R., Seaman C., Sullivan K., Zazworka N.,

2010. Managing technical debt in software-reliant sys-

tems. Proceedings of the FSE/SDP workshop on Future

of software engineering research, pp. 47 – 52, ACM.

Buschmann F., 2011. To Pay or Not to Pay Technical Debt.

Software, 28 (6), pp. 29-31, IEEE Computer Society.

Chin S., Huddleston E., Bodwell W., Gat I., 2010. The Eco-

nomics of Technical Debt, Cutter IT Journal.

Codabux Z., Williams B., 2013. Managing technical debt:

An industrial case study. 4

International Workshop on

Managing Technical Debt (MTD ‘13). IEEE Computer

Society.

Cunningham W., 1992. The WyCash Portfolio Manage-

ment System. 7

International Conference on Object-

Oriented Programming, Systems, Languages, and Ap-

plications (OOPSLA ’92).

Curtis B., Sappidi J., Szynkarski A., 2012. Estimating the

Principal of an Application's Technical Debt. Software

29 (6), pp. 34 - 42, IEEE Computer Society.

Curtis B., Sappidi J., Szynkarski A., 2012. Estimating the

size cost and types of Technical Debt. 3

International

Workshop on Managing Technical Debt (MTD ‘12).

IEEE Computer Society.

Eisenberg R. J., 2012. A threshold based approach to tech-

nical debt. ACM SIGSOFT Software Engineering

Notes, 37 (2), pp. 1 - 6, ACM.

Ernst N., 2012. On the role of requirements in understand-

ing and managing technical debt. 3

International

Workshop on Managing Technical Debt (MTD ‘12).

IEEE Computer Society.

Falessi D., Shaw M., Shull F., Mullen K., Keymind M.,

2013. Practical considerations challenges and require-

ments of tool-support for managing technical debt. 4

International Workshop on Managing Technical Debt

(MTD ‘13). IEEE Computer Society.

de Groot J., Nugroho A., Back T., Visser J., 2012. What is

the value of your software?, 3

International Workshop

on Managing Technical Debt (MTD ‘12). IEEE Com-

puter Society.

Guo Y., Seaman C., 2011. A portfolio approach to technical

debt management. 2

International Workshop on Man-

aging Technical Debt. ACM.

Guo Y., Seaman C., Gomes R., Cavalcanti A., Tonin G., da

Silva F., Santos A.L., Siebra C., 2011. Tracking tech-

nical debt - An exploratory case study. 27

Interna-

tional Conference on Software Maintenance (ICSM

‘11). IEEE Computer Society.

Holvitie J., Leppanen V., 2013. DebtFlag: Technical debt

management with a development environment inte-

grated tool. 4

International Workshop on Managing

Technical Debt (MTD ‘13). IEEE Computer Society.

Kitchenham B., Brereton O. P., Budgen D., Turner M., Bai-

ley J., Linkman S., 2009. Systematic literature reviews

in software engineering – A systematic literature re-

view. Information and Software Technology, 51 (1), pp

7-15. Elsevier.

Koolmanojwongand S., Lane J.A., 2013. Enablers and In-

hibitors of Expediting Systems Engineering. 11

An-

nual Conference on Systems Engineering Research

(CSER ‘13). Procedia Computer Science, Elsevier.

Li Z., Avgeriou P., Liang P., 2015. A systematic mapping

study on technical debt and its management. Journal of

Systems and Software, 101, pp. 193-220. Elsevier.

Letouzey J. L., 2012. The sqale method for evaluating tech-

nical debt. 3

International Workshop on Managing

Technical Debt (MTD ‘12). IEEE Computer Society.

Marinescu R., 2012. Assessing technical debt by identify-

ing design flaws in software systems. Journal of Re-

search and Development, 56 (5), pp. 1-13, IBM.

McGregor J. D., Monteith J., Zhang J., 2012. Technical

debt aggregation in ecosystems. 3

International Work-

shop on Managing Technical Debt (MTD ‘12). IEEE

Computer Society.

Mishkin F., Eakins S., 2012. Financial Markets and Insti-

tutions. Pearson Prentice Hall, 7

Edition.

Nord R., Ozkaya I., Kruchten P., Gonzalez-Rojas M., 2012.

In Search of a Metric for Managing Architectural Tech-

nical Debt. 2012 Joint Working IEEE/IFIP Conference

on Software Architecture (WICSA) and European Con-

ference on Software Architecture (ECSA). IEEE Com-

puter Society.

Nugroho A., Visser J., Kuipers T., 2011. An empirical

model of technical debt and interest. 2nd International

Workshop on Managing Technical Debt (MTD’ 11).

ACM.

Schmid K., 2013. On the limits of the technical debt meta-

phor some guidance on going beyond. 4

International

Workshop on Managing Technical Debt (MTD ‘13).

IEEE Computer Society.

Seaman C., Guo Y., 2011. Measuring and monitoring tech-

nical debt. Advances in Computers, 82, pp. 25-46. Else-

vier.

Seaman C., Guo Y., Zazworka N., Shull F., Izurieta C., Cai

Y., Vetro A., 2012. Using technical debt data in deci-

sion making: Potential decision approaches. 3rd Inter-

national Workshop on Managing Technical Debt

(MTD’ 12). IEEE Computer Society.

Siebra C. S., Tonin G. S., Silva F. Q., Oliveira R. G., Junior

A. L., Miranda R. C., Santos A. L., 2012. Managing

technical debt in practice: an industrial report. 6th In-

ternational Symposium on Empirical Software Engi-

neering and Measurement (ESEM’ 12). ACM.

Snipes W., Robinson B., Guo Y., Seaman C., 2012. Defin-

ing the decision factors for managing defects: A tech-

nical debt perspective. 3rd International Workshop on

Managing Technical Debt (MTD’ 12). IEEE Computer

Society.

Fifth International Symposium on Business Modeling and Software Design

Zazworka N., Seaman C., Shull F., 2011. Prioritizing de-

sign debt investment opportunities. 2

International

Workshop on Managing Technical Debt (MTD’ 11).

ACM.

Zazworka N., Spanola R. O., Vetró A., Shull F. Seaman

C., 2013. A case study on effectively identifying tech-

nical debt. 17

International Conference on Evaluation

and Assessment in Software Engineering. ACM.

Zazworka N., Vetró A., Izurieta C., Wong S., Cai Y., Sea-

man C. Shull F., 2014. Comparing four approaches for

technical debt identification. Software Quality Journal.

22 (3), pp. 403 – 426, Springer.

Establishing a Framework for Managing Interest in Technical Debt